diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..8a04a4c16caa346adadbaf0442dc56d53516f4e5 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -33,3 +33,121 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+benchmark/banner/image/\#3011374224600092738.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/banner/image/\#5052752366753654.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/banner/image/\#7201809802947156804.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/banner/image/\#823283996711975890.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/banner/image/\#905615867444667299.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/banner/image/\#9191033322170410315.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/banner/image/4203027269594344041.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/6GwqzUtcLIiiQVynsOsIe.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/6IJlRrwHnU2DNAHnhWbQI.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/7QLIsS96jFQnvBid9KDvt.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/A3dOllegcWLDe55BGf4iT.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/E9TszYni5Gu1JDkoK-k2x.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/dm8WI-nJXmpN-kqg0C6gX.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/e8bCv7DetdVCCHV3-zRl3.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/ep1OYqiShiauTYnBt3jmV.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/kZ8pPPinTrxbL42ECZJ5J.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/n3i0yF2iBMpIP8RbzJg3A.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/rLw9t28WNqJP5OnIrWVge.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/u0CwaM61V6TeawWryt0DC.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car/image/vB4uib2oeDwoizQD2oE5T.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/\#2563290238964912476.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/\#6227411411065045336.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/\#761631743535467855.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/1794121684277643246.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/3021668729827686673.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/318568126668983171.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/3488762191409319384.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/car_plate/image/6386760609565435452.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_22_17_04_01_726-fishing.mp4\#\#76.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_23_05_59_27_157-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_23_07_33_18_206-fishing.mp4\#\#163.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_23_08_41_12_169-fishing.mp4\#\#245.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_23_11_21_29_168-fishing.mp4\#\#217.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_23_12_31_33_969-fishing.mp4\#\#162.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_23_16_34_50_593-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_23_18_34_53_717-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_24_06_13_29_399-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_25_09_16_12_276-fishing.mp4\#\#158.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_25_09_33_38_445-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_25_13_56_51_638-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_25_17_03_08_137-fishing.mp4\#\#169.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_26_09_38_00_142-fishing.mp4\#\#70.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_26_10_41_23_352-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_26_10_53_31_282-fishing.mp4\#\#74.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_04_26_11_04_01_928-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_01_08_50_50_652-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_07_08_38_51_400-fishing.mp4\#\#88.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_08_10_25_40_742-fishing.mp4\#\#137.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_08_11_53_34_002-fishing.mp4\#\#69.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_08_12_39_58_887-fishing.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_08_15_23_31_150-fishing.mp4\#\#75.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_11_06_30_17_609-fishing.mp4\#\#129.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_17_08_55_21_721-fishing.mp4\#\#80.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_05_17_09_28_14_597-fishing.mp4\#\#145.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_06_19_13_01_47_221-fishing.mp4\#\#34.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#changjiang\#\#2021_06_24_15_42_29_196-fishing.mp4\#\#62.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#chowder\#\#江堤达标纪念碑东侧高倍球机_1600249854808.mp4\#\#102.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/image/image\#\#chowder\#\#浒通河闸西侧普通球机_1600327196011.mp4\#\#1.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/fisherman/参考图片/垂钓人员检测20251026_130031_3.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#1026.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#103.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#1412.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#153.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#205.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#246.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#27.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#279.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#354.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#403.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#45.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#53.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#533.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#562.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#596.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#626.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#647.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#651.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#663.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#68.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#681.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#766.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#842.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#901.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#902.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#911.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#934.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/image/image\#\#images\#\#960.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/garbage_dump/参考图片/image\#\#images\#\#153.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/ped_bev/image/image\#\#dazuzhuang\#\#3_0180.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/-xGxHXYw3G6VvF0X6LHzM.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/BGP8kQvB9XhKZeHfRoIkL.jpeg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/Va4djcAIDTBHxkEWh3sR9.jpeg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/ZgOaHSLAgWggjuMS9GPHI.jpeg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/dbkdbFKii3AY1WB9Ddn3g.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/km8YIqcvHVZTvZm2BPzDN.jpeg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/lJqnoPxfepG7vPyL8wcEE.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/liBRM7GLZTTnNr43skYCk.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/r4wgU8UN8zvdxgOzCDJrh.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/tuWVK9G0eu9dacLrElQbD.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/wizgEUAJ5GulK13NITOu7.jpeg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/x_2joRmo8KHRQfveORsMh.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/smog/image/yDDmmcEzFjJG62qvVh8oh.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/-HriUP8AIMoxb6ARMUqe_.PNG filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/54QUfL1zu9KxdgC0Lnkaq.PNG filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/5PfcirXJfTg3STKLMXt3Z.PNG filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/6VsaAYiP1DGftITp0-CMc.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/9Tq31VYy_-FXffXrf8wrC.PNG filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/9zQR3UTnc106GQMYliRIf.webp filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/AFPldc8N0Wt6VqquthNYc.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/CzNfn-KcVwwT94-0jKuvH.png filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/D3l_pT9SoB5NrTPJHey8e.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/DF6-H_sErOB8BgG6VDcgj.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/F9Jp_THhd1JDVHDwIfyvU.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/I9tzSGkcPQqvg9iKeJmxH.PNG filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/IS_8jyoAarGlGjEQPIMBt.PNG filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/IkKMCU1PTdFGeAclmetow.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/J5MeAlZD__Sq5QcJyf4gv.jpg filter=lfs diff=lfs merge=lfs -text
+benchmark/water/image/Ji5jAGAaWpGp7l5nPgfCa.PNG filter=lfs diff=lfs merge=lfs -text
diff --git a/benchmark/banner/gt.jsonl b/benchmark/banner/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..b8c067470232f6f0bb451afb4acc4adbfde057ce
--- /dev/null
+++ b/benchmark/banner/gt.jsonl
@@ -0,0 +1,20 @@
+{"filename": "image/#9087706379781072719.jpg", "image_height": 300, "image_width": 484, "instances": [{"is_ignored": false, "bbox": [52, 11, 108, 66], "label": 1}, {"is_ignored": false, "bbox": [201, 44, 272, 110], "label": 1}, {"is_ignored": false, "bbox": [263, 90, 348, 174], "label": 1}, {"is_ignored": false, "bbox": [208, 116, 258, 170], "label": 1}, {"is_ignored": false, "bbox": [83, 74, 152, 124], "label": 1}, {"is_ignored": false, "bbox": [233.1311, 158.4833, 429.6959, 300.0], "label": 1}, {"is_ignored": false, "bbox": [452.4488, 155.3169, 484.0, 298.57529999999997], "label": 1}, {"is_ignored": false, "bbox": [429.696, 164.1653, 452.02750000000003, 270.3451], "label": 1}, {"is_ignored": false, "bbox": [440.651, 102.227, 469.724, 122.0304], "label": 1}, {"is_ignored": false, "bbox": [366.4937, 48.2945, 405.6791, 81.58099999999999], "label": 1}, {"is_ignored": false, "bbox": [337.4207, 100.5417, 359.3308, 126.24390000000001], "label": 1}, {"is_ignored": false, "bbox": [152.0274, 71.89, 192.8982, 107.2833], "label": 1}, {"is_ignored": false, "bbox": [41.6342, 56.3001, 74.9207, 90.4293], "label": 1}, {"is_ignored": false, "bbox": [1.1847, 124.1372, 23.9375, 149.8394], "label": 1}]}
+{"filename": "image/4203027269594344041.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1011.6378, 69.7921, 1317.8555999999999, 105.59280000000001], "label": 1}, {"is_ignored": false, "bbox": [1025.7932, 54.3786, 1277.7318, 91.19149999999999], "label": 1}]}
+{"filename": "image/500460508410305335.jpg", "image_height": 198, "image_width": 300, "instances": [{"is_ignored": false, "bbox": [241, 158, 298, 197], "label": 1}, {"is_ignored": false, "bbox": [140.7811, 135.0203, 168.0291, 150.0904], "label": 1}, {"is_ignored": false, "bbox": [260.195, 110.4508, 299.7503, 155.3025], "label": 1}]}
+{"filename": "image/#7100140247854464792.jpg", "image_height": 325, "image_width": 490, "instances": [{"is_ignored": false, "bbox": [331, 59, 444, 135], "label": 1}, {"is_ignored": false, "bbox": [192, 182, 240, 233], "label": 1}]}
+{"filename": "image/#8599663314952454005.jpg", "image_height": 300, "image_width": 399, "instances": [{"is_ignored": false, "bbox": [40, 150, 251, 298], "label": 1}]}
+{"filename": "image/#9191033322170410315.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1655.6881, 48.741, 1784.9652, 81.4435], "label": 1}]}
+{"filename": "image/5625359483858974892.jpg", "image_height": 375, "image_width": 500, "instances": [{"is_ignored": false, "bbox": [140, 48, 361, 274], "label": 1}]}
+{"filename": "image/#905615867444667299.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [104.0832, 91.9267, 198.5749, 111.6362], "label": 1}]}
+{"filename": "image/#823283996711975890.jpg", "image_height": 450, "image_width": 800, "instances": [{"is_ignored": false, "bbox": [123, 159, 799, 345], "label": 1}]}
+{"filename": "image/#228241907577826423.jpg", "image_height": 300, "image_width": 465, "instances": [{"is_ignored": false, "bbox": [170, 180, 228, 224], "label": 1}, {"is_ignored": false, "bbox": [207.2015, 118.2115, 251.08210000000003, 147.7637], "label": 1}, {"is_ignored": false, "bbox": [299.8881, 69.8533, 332.1269, 97.1667], "label": 1}, {"is_ignored": false, "bbox": [375.1119, 68.9578, 401.52979999999997, 89.5548], "label": 1}, {"is_ignored": false, "bbox": [428.3955, 51.0473, 465.0, 97.61449999999999], "label": 1}, {"is_ignored": false, "bbox": [52.2761, 91.7936, 79.1418, 131.6443], "label": 1}, {"is_ignored": false, "bbox": [176.7537, 30.4503, 219.73880000000003, 57.7637], "label": 1}, {"is_ignored": false, "bbox": [183.9179, 65.8234, 200.03730000000002, 80.15180000000001], "label": 1}]}
+{"filename": "image/#2106501120964560567.jpg", "image_height": 332, "image_width": 500, "instances": [{"is_ignored": false, "bbox": [87, 29, 230, 153], "label": 1}, {"is_ignored": false, "bbox": [48, 58, 97, 143], "label": 1}, {"is_ignored": false, "bbox": [230.4269, 0, 316.64779999999996, 100.0983], "label": 1}, {"is_ignored": false, "bbox": [45.597, 0, 79.78800000000001, 44.1043], "label": 1}]}
+{"filename": "image/#5052752366753654.jpg", "image_height": 395, "image_width": 770, "instances": [{"is_ignored": false, "bbox": [0, 96, 769, 191], "label": 1}]}
+{"filename": "image/#3011374224600092738.jpg", "image_height": 458, "image_width": 690, "instances": [{"is_ignored": false, "bbox": [97, 333, 685, 425], "label": 1}, {"is_ignored": false, "bbox": [0, 216, 665, 326], "label": 1}, {"is_ignored": false, "bbox": [0, 304.7572, 267.6961, 351.4214], "label": 1}]}
+{"filename": "image/#5558124080853495957.jpg", "image_height": 360, "image_width": 640, "instances": [{"is_ignored": false, "bbox": [198.0559, 182.26, 640.0, 322.5946], "label": 1}]}
+{"filename": "image/1013182005258676588.jpg", "image_height": 313, "image_width": 500, "instances": [{"is_ignored": false, "bbox": [25, 165, 338, 311], "label": 1}, {"is_ignored": false, "bbox": [189, 235, 500, 311], "label": 1}, {"is_ignored": false, "bbox": [51, 0, 136, 136], "label": 1}, {"is_ignored": false, "bbox": [305, 0, 371, 79], "label": 1}, {"is_ignored": false, "bbox": [186, 0, 295, 171], "label": 1}, {"is_ignored": false, "bbox": [404, 0, 499, 131], "label": 1}, {"is_ignored": false, "bbox": [107, 0, 218, 57], "label": 1}, {"is_ignored": false, "bbox": [307, 29, 398, 138], "label": 1}]}
+{"filename": "image/4486337562479806726.jpg", "image_height": 437, "image_width": 616, "instances": [{"is_ignored": false, "bbox": [213, 222, 569, 334], "label": 1}]}
+{"filename": "image/6157173247889961715.jpg", "image_height": 435, "image_width": 580, "instances": [{"is_ignored": false, "bbox": [0, 90.261, 575.9237, 295.2409], "label": 1}]}
+{"filename": "image/6277895318465160012.jpg", "image_height": 310, "image_width": 500, "instances": [{"is_ignored": false, "bbox": [394, 79, 452, 139], "label": 1}, {"is_ignored": false, "bbox": [296, 59, 359, 127], "label": 1}, {"is_ignored": false, "bbox": [187, 82, 230, 145], "label": 1}, {"is_ignored": false, "bbox": [26, 143, 340, 285], "label": 1}]}
+{"filename": "image/7128298139195245589.jpg", "image_height": 315, "image_width": 500, "instances": [{"is_ignored": false, "bbox": [146, 161, 499, 313], "label": 1}, {"is_ignored": false, "bbox": [0, 175, 210, 313], "label": 1}, {"is_ignored": false, "bbox": [203, 57, 269, 103], "label": 1}, {"is_ignored": false, "bbox": [0, 156.7438, 43.6925, 227.3167], "label": 1}]}
+{"filename": "image/9016722135656931961.jpg", "image_height": 333, "image_width": 500, "instances": [{"is_ignored": false, "bbox": [216, 55, 390, 256], "label": 1}, {"is_ignored": false, "bbox": [129.6348, 226.5167, 159.7623, 248.37669999999997], "label": 1}, {"is_ignored": false, "bbox": [196.6159, 221.4721, 218.7561, 242.6314], "label": 1}]}
diff --git a/benchmark/banner/image/#2106501120964560567.jpg b/benchmark/banner/image/#2106501120964560567.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b4f4b2770a55f90607020d1baf92a06a1a0ccf62
Binary files /dev/null and b/benchmark/banner/image/#2106501120964560567.jpg differ
diff --git a/benchmark/banner/image/#228241907577826423.jpg b/benchmark/banner/image/#228241907577826423.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b6b1fe55455742c4156abc8a923cb0609ceca00c
Binary files /dev/null and b/benchmark/banner/image/#228241907577826423.jpg differ
diff --git a/benchmark/banner/image/#3011374224600092738.jpg b/benchmark/banner/image/#3011374224600092738.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4f15300c2396fd5e12ddb0017c09f19dd67a65ce
--- /dev/null
+++ b/benchmark/banner/image/#3011374224600092738.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e3999f2e9571ca63aeef5cf25d2860adbdddbaa7e064f2b9e3775d2db0b299b
+size 186825
diff --git a/benchmark/banner/image/#5052752366753654.jpg b/benchmark/banner/image/#5052752366753654.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4814bbc4bc26545d63d17503462a94c750d0910e
--- /dev/null
+++ b/benchmark/banner/image/#5052752366753654.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:803e8ced95443f8b6292f7077c914645fcb2111163487f9230d29b92678282a6
+size 174306
diff --git a/benchmark/banner/image/#5558124080853495957.jpg b/benchmark/banner/image/#5558124080853495957.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..0f3b0d72ec008e0724f7b3ae5d46ababb780bdf9
Binary files /dev/null and b/benchmark/banner/image/#5558124080853495957.jpg differ
diff --git a/benchmark/banner/image/#7100140247854464792.jpg b/benchmark/banner/image/#7100140247854464792.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b4a6d33e1d1a2fda2cb681a939753fa6a36b1839
Binary files /dev/null and b/benchmark/banner/image/#7100140247854464792.jpg differ
diff --git a/benchmark/banner/image/#7201809802947156804.jpg b/benchmark/banner/image/#7201809802947156804.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..88269c88f16437a0f3076875c710d2c9c6f22367
--- /dev/null
+++ b/benchmark/banner/image/#7201809802947156804.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:129a87befae21ec018bfa45e4cbf86df0c12f579c566ba13d2608a7fff0b0770
+size 430697
diff --git a/benchmark/banner/image/#823283996711975890.jpg b/benchmark/banner/image/#823283996711975890.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..87799b5a916e4f84531b3b6c4ea84d8179c6c30c
--- /dev/null
+++ b/benchmark/banner/image/#823283996711975890.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fab6c6a3868dd6caa060a39f0e9cd740fbad9d883658cb0ef71e1628ab68dd9
+size 156611
diff --git a/benchmark/banner/image/#8599663314952454005.jpg b/benchmark/banner/image/#8599663314952454005.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..fba9abd02de96b3348e2fd374e2eba4453f471b4
Binary files /dev/null and b/benchmark/banner/image/#8599663314952454005.jpg differ
diff --git a/benchmark/banner/image/#905615867444667299.jpg b/benchmark/banner/image/#905615867444667299.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..61c1dcdfcc02519be6d46667400f86a733b76e18
--- /dev/null
+++ b/benchmark/banner/image/#905615867444667299.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a12af92c13373962d7064770fe229f6b4704f9dc03ff7e04d620e995bec2060a
+size 165041
diff --git a/benchmark/banner/image/#9087706379781072719.jpg b/benchmark/banner/image/#9087706379781072719.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..36d03bdfc47ae8a452a903f1bce676fb44eeb790
Binary files /dev/null and b/benchmark/banner/image/#9087706379781072719.jpg differ
diff --git a/benchmark/banner/image/#9191033322170410315.jpg b/benchmark/banner/image/#9191033322170410315.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8e8bacb804229c7ab8ef9ecc218fbd327a76077c
--- /dev/null
+++ b/benchmark/banner/image/#9191033322170410315.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f3fa6f44061dfdb590236fe9254ab19d66d8a7780ccc045bd084c7d1a7b359c
+size 671122
diff --git a/benchmark/banner/image/1013182005258676588.jpg b/benchmark/banner/image/1013182005258676588.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..bb4d821d4cd3c6649550902b3de0da066d42a6da
Binary files /dev/null and b/benchmark/banner/image/1013182005258676588.jpg differ
diff --git a/benchmark/banner/image/4203027269594344041.jpg b/benchmark/banner/image/4203027269594344041.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3e712e6f54db25ad44a60de1deabe45a3a0ce5a3
--- /dev/null
+++ b/benchmark/banner/image/4203027269594344041.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87f711c0a4c0801f73e95c4b6b048021720b41c858e8064327ad2ad12e1a8c86
+size 566684
diff --git a/benchmark/banner/image/4486337562479806726.jpg b/benchmark/banner/image/4486337562479806726.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..10ab042c4af5f9385265c7978760d797b468aa4b
Binary files /dev/null and b/benchmark/banner/image/4486337562479806726.jpg differ
diff --git a/benchmark/banner/image/500460508410305335.jpg b/benchmark/banner/image/500460508410305335.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ea3ab5bd6c3beedb06b6ede9de30d7c416a14c7f
Binary files /dev/null and b/benchmark/banner/image/500460508410305335.jpg differ
diff --git a/benchmark/banner/image/5625359483858974892.jpg b/benchmark/banner/image/5625359483858974892.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..dc8134a3fe271c732b38aa1e29a93e6e64456afb
Binary files /dev/null and b/benchmark/banner/image/5625359483858974892.jpg differ
diff --git a/benchmark/banner/image/6157173247889961715.jpg b/benchmark/banner/image/6157173247889961715.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..9c33b837c657981f6a2a767f403fbcf903374dd9
Binary files /dev/null and b/benchmark/banner/image/6157173247889961715.jpg differ
diff --git a/benchmark/banner/image/6277895318465160012.jpg b/benchmark/banner/image/6277895318465160012.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..c35db0430b2a83327e6c79bf797683ab0ee17537
Binary files /dev/null and b/benchmark/banner/image/6277895318465160012.jpg differ
diff --git a/benchmark/banner/image/7128298139195245589.jpg b/benchmark/banner/image/7128298139195245589.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4d35e539ebd5efb65230aa3209361cabc569fab8
Binary files /dev/null and b/benchmark/banner/image/7128298139195245589.jpg differ
diff --git a/benchmark/banner/image/9016722135656931961.jpg b/benchmark/banner/image/9016722135656931961.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..20df15bea52698795f565b06f37f51b49f13af4c
Binary files /dev/null and b/benchmark/banner/image/9016722135656931961.jpg differ
diff --git a/benchmark/banner/image_description.jsonl b/benchmark/banner/image_description.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..a21b7a9f0646ad578d4d43843be91202eca8eae4
--- /dev/null
+++ b/benchmark/banner/image_description.jsonl
@@ -0,0 +1,20 @@
+{"image_path": "image/#9087706379781072719.jpg", "image_description": "The image shows a group of people participating in a protest. They are holding signs and banners with various messages. The most prominent sign in the foreground reads \"NO! MD!\" in bold letters on a green background. Other signs include phrases like \"NO!! THAAD,\" \"AN-1,\" and \"SPARK.\" The participants appear to be expressing opposition to something, possibly related to military defense systems or missile defense, as suggested by the text on the signs.\nThe setting appears to be outdoors, possibly near a government building or a public space, given the presence of trees and a fence in the background. The atmosphere seems serious and focused, indicating that the protest is likely about an important issue."}
+{"image_path": "image/4203027269594344041.jpg", "image_description": "The image shows an outdoor scene at what appears to be a school or public building on a rainy day. There are several people holding umbrellas, and they are standing in a queue under a red canopy with a sign that reads \"520号\" (indicating a number, possibly for a specific purpose like a registration or check-in process). The ground is wet from the rain, and there are barriers set up to manage the flow of people.\nIn the background, there is a building with a red facade and arched entrances. Red lanterns are hanging above the entrance, suggesting a festive or celebratory occasion. The timestamp on the image indicates it was taken on November 25, 2020, at 15:24:52.\nRegarding the target banner, there is a banner visible in the background near the entrance of the building. It is partially obscured but appears to have text on it, which might be related to the event or occasion taking place. The text on the banner is not fully legible due to the angle and distance, but it seems to be in Chinese characters."}
+{"image_path": "image/500460508410305335.jpg", "image_description": "The image depicts a dynamic scene with a person in mid-air, seemingly performing a jump or a flip. The individual appears to be wearing casual clothing and is positioned above a crowd of people who are watching the action unfold. The background features a row of buildings with a traditional architectural style, suggesting an urban setting.\nRegarding the target banner, there is no visible banner in the image that matches the description provided. If you have any specific details about the banner you're referring to, please provide them for further assistance."}
+{"image_path": "image/#7100140247854464792.jpg", "image_description": "The image shows three individuals standing on a rooftop under a clear blue sky. They appear to be dressed in costumes and are holding signs with text on them. The person in the center is wearing a mask and has a sign that reads \"FUKING KILLERS.\" The other two individuals are also wearing masks and holding signs, but the text on their signs is not fully visible or legible in the image. The setting suggests a public event or protest, given the nature of the signs and the attire of the participants."}
+{"image_path": "image/#8599663314952454005.jpg", "image_description": "The image shows a person wearing a white protective suit with a hood and holding a sign that reads \"STOP!\" in large red letters. The sign also has some additional text in what appears to be Japanese characters below the English word \"STOP.\" The person is also holding a small object resembling a frying pan with a radiation symbol on it. The background includes other individuals holding similar signs and banners, suggesting a protest or demonstration. The setting appears to be outdoors under a clear blue sky."}
+{"image_path": "image/#9191033322170410315.jpg", "image_description": "The image depicts a busy urban street scene in Guangzhou, China, as indicated by the text on the right side of the image. The timestamp on the image shows it was taken on November 4, 2018, at 14:35:26.\n### Description of the Image:\n- **Street and Traffic:** The street is lined with various shops and restaurants on the left side, while vehicles are moving along the road on the right. There are several cars and a truck visible.\n- **Pedestrians:** There are people walking on the sidewalk, some carrying bags or pushing strollers. The pedestrians appear to be going about their daily activities.\n- **Buildings and Shops:** The buildings have signs in Chinese characters, indicating they are likely commercial establishments. One prominent sign reads \"巴黎春天\" (Paris Spring), which could be a shop name.\n- **Sidewalk and Infrastructure:** The sidewalk is paved, and there are bollards placed at the corner for safety. A white railing separates the sidewalk from the road.\n- **Weather and Lighting:** The weather appears to be overcast, as the sky is not very bright, and the lighting suggests it might be late afternoon or early evening.\n### Target Banner:\nThere is no specific target banner identified in the image. However, there are multiple signs and banners visible on the buildings, including one that says \"巴黎春天\" (Paris Spring) and another that appears to be an advertisement for a restaurant or food service. These signs are part of the general signage on the buildings and do not seem to be a target banner in the context of the question."}
+{"image_path": "image/5625359483858974892.jpg", "image_description": "The image shows a group of people gathered outdoors, possibly at a protest or demonstration. In the foreground, a person is holding up a sign that reads \"FREE BRADLEY MANNING!\" The sign is white with black text and appears to be handmade. The background features modern buildings, including a tall structure with a distinctive design and another building with a curved facade. The atmosphere suggests a public gathering with a focus on a specific cause or individual."}
+{"image_path": "image/#905615867444667299.jpg", "image_description": "The image depicts a busy urban scene with several people walking and a red bus labeled \"325\" parked on the right side. The bus has advertisements on its body. There is also a yellow vehicle partially visible behind the bus. On the left side, there are more people walking, some carrying bags or pushing strollers. The ground appears wet, suggesting recent rain.\nIn the foreground, there is a small kiosk with various items displayed, including what looks like snacks or drinks. A blue railing runs along the edge of the kiosk area. The timestamp in the top right corner indicates the date and time as \"2018-12-27 09:51:19.\"\nRegarding the target banner, it is not clearly visible in the provided image. If you have a specific target banner in mind, please provide more details or clarify which one you are referring to."}
+{"image_path": "image/#823283996711975890.jpg", "image_description": "The image shows a staircase leading up to a building with a red banner stretched across it. The banner has yellow Chinese characters that read \"人人都爱太子妃 建院女神你最美,\" which translates to \"Everyone loves Princess Taisi, you are the most beautiful goddess at the hospital.\" The background includes a building with beige walls and some greenery, including palm trees and other plants. The bottom right corner of the image has a logo and text indicating the source as \"人民网 福建频道\" (People's Daily Online Fujian Channel)."}
+{"image_path": "image/#228241907577826423.jpg", "image_description": "The image depicts a large gathering of people participating in what appears to be a protest or demonstration. The crowd is densely packed, with many individuals holding signs and banners. The atmosphere suggests a peaceful assembly, possibly advocating for a cause.\nIn the center of the image, there is a prominent figure wearing a yellow hat and a mask that resembles a cartoonish face. This figure is holding a sign that reads \"VIVA LA MUJER,\" which translates to \"Long Live Women\" in English. The sign features a symbol resembling a female gender sign with a heart shape inside it, emphasizing the theme of women's rights and empowerment.\nThe crowd around this central figure is diverse, with many participants wearing pink accessories, which could be symbolic of various causes such as women's rights, LGBTQ+ rights, or other social justice movements. The overall scene conveys a sense of unity and solidarity among the attendees."}
+{"image_path": "image/#2106501120964560567.jpg", "image_description": "The image depicts a large crowd gathered outdoors, likely at a public event or demonstration. The crowd appears to be diverse, with many individuals holding flags and signs. A prominent rainbow flag is visible, suggesting that the event may have LGBTQ+ themes.\nIn the foreground, there are several police officers wearing uniforms with \"POLICJA\" written on their backs, indicating they are from the Polish police force. One of the officers is holding a sign that reads \"TU JEST POLSKA,\" which translates to \"HERE IS POLAND\" in English. This suggests that the event might be related to national pride or identity.\nThe atmosphere seems lively and energetic, with people engaged in conversation and activities. The setting appears to be an urban area, possibly a street or square, given the presence of buildings and infrastructure in the background."}
+{"image_path": "image/#5052752366753654.jpg", "image_description": "The image shows a brick wall with a red banner stretched across it. The banner has white Chinese characters written on it. Below the banner, there is a table with some items on it, including what appears to be a black bag and a cup. There is also a poster or flyer attached to the wall below the banner, featuring an illustration of flowers and some text. The overall setting seems to be outdoors, possibly near a school or institution given the context of the banner and the poster."}
+{"image_path": "image/#3011374224600092738.jpg", "image_description": "The image shows an outdoor scene with a multi-story residential building in the background. There are leafless trees in front of the building, suggesting it might be late autumn or winter. In the foreground, there are two red banners stretched across the street. The text on the banners is written in Chinese characters.\nThe top banner has white text that reads: \"今晚请你吃啤酒和炸鸡\" which translates to \"Tonight, I will treat you to beer and fried chicken.\" Below this, there is smaller text that appears to be additional information or a note, but it is not fully legible due to the angle and resolution.\nThe bottom banner also has white text that reads: \"环11女神，男仆们已经准备好啤酒炸鸡了!\" This translates to \"Ring 11 Goddess, the male servants have already prepared beer and fried chicken!\"\nThe banners seem to be part of an event or promotion, possibly related to a celebration or gathering involving the mentioned \"Ring 11 Goddess\" and her male companions."}
+{"image_path": "image/#5558124080853495957.jpg", "image_description": "The image shows an urban scene with a red banner prominently displayed on a metal railing. The banner has white Chinese characters that read \"网络交友须谨\" which translates to \"Caution when making online friends.\" The background features modern high-rise buildings and some greenery. There is also a yellow warning sign on a black post near the banner, indicating potential hazards or restrictions. The overall setting appears to be a public area, possibly a pedestrian walkway or a bridge."}
+{"image_path": "image/1013182005258676588.jpg", "image_description": "The image shows a group of people participating in what appears to be a protest or demonstration. They are holding signs and flags, with one prominent sign reading \"TÜRKİYE SUR SAVAŞ\" (Turkey Sur War) in yellow text on a yellow background. The flags include one with a black and white portrait of a person and another with red and white text. The participants are raising their fists, suggesting a show of solidarity or defiance. The setting seems to be outdoors, possibly in a park or public square."}
+{"image_path": "image/4486337562479806726.jpg", "image_description": "The image shows a rural scene with a stone wall in the foreground and some buildings in the background under a partly cloudy sky. There is a red banner hanging on the stone wall with Chinese text that translates to \"Precise identification, precise measures, precise poverty alleviation and eradication\" and \"Zhongshan City驻村工作队宣,\" which indicates it is related to a poverty alleviation program by a work team from Zhongshan City. The banner is prominently displayed and appears to be part of an initiative aimed at addressing poverty in the area."}
+{"image_path": "image/6157173247889961715.jpg", "image_description": "The image shows a red banner hanging across a green bush in front of a building with white walls and multiple windows. The text on the banner is written in Chinese characters. The banner reads \"林工的帅哥们，男生节快乐\" which translates to \"Happy Boys' Day, handsome guys from Lining.\" The background includes a tree trunk on the left side and part of the building's facade."}
+{"image_path": "image/6277895318465160012.jpg", "image_description": "The image shows a group of people participating in what appears to be a protest or demonstration. The individuals are holding signs and banners, with one prominent red banner that reads \"NO PASARÁN\" (Spanish for \"They shall not pass\") and \"RAGE AGAINST FASCISM.\" The text on the banner is bold and black, making it stand out against the red background.\nThe scene takes place in an urban area, likely a city street, with buildings and storefronts visible in the background. The atmosphere suggests a peaceful protest, as there are no visible signs of conflict or violence. The participants appear to be wearing casual clothing, and some are holding up signs with Japanese characters, indicating that the event might be taking place in Japan."}
+{"image_path": "image/7128298139195245589.jpg", "image_description": "The image shows a group of people participating in what appears to be a protest or demonstration. They are holding signs and banners with various messages. The most prominent banner in the foreground has text in Vietnamese and English. The Vietnamese text reads \"Đoàn kết dân tộc là sức mạnh\" which translates to \"Unity of the nation is strength.\" Below that, it says \"Vietnam! United we stand, divided we fall,\" which is a common slogan used in protests to emphasize the importance of unity.\nThe participants are wearing casual clothing, and some are raising their fists in solidarity. The background includes urban elements such as buildings, trees, and other protesters, indicating an outdoor setting. The overall atmosphere suggests a peaceful yet passionate demonstration."}
+{"image_path": "image/9016722135656931961.jpg", "image_description": "The image shows a nighttime scene with a group of people gathered outside a building that has \"TRUMP TOWER\" written on it. The focus is on a person holding up a sign that reads \"NOT MY PRESIDENT\" in bold, black letters on a white background. The sign is held high above the crowd, and the background includes the illuminated windows of Trump Tower and some streetlights. The atmosphere suggests a protest or demonstration."}
diff --git a/benchmark/banner/user_input.txt b/benchmark/banner/user_input.txt
new file mode 100644
index 0000000000000000000000000000000000000000..53e8c9c5e74e1b9ff2823c0027c266fc43f4936b
--- /dev/null
+++ b/benchmark/banner/user_input.txt
@@ -0,0 +1 @@
+检测旗帜横幅
\ No newline at end of file
diff --git "a/benchmark/banner/\345\217\202\350\200\203\345\233\276\347\211\207/4486337562479806726.jpg" "b/benchmark/banner/\345\217\202\350\200\203\345\233\276\347\211\207/4486337562479806726.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..10ab042c4af5f9385265c7978760d797b468aa4b
Binary files /dev/null and "b/benchmark/banner/\345\217\202\350\200\203\345\233\276\347\211\207/4486337562479806726.jpg" differ
diff --git a/benchmark/car/gt.jsonl b/benchmark/car/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..d4650687c0d9512a6073e85716d11be1a6427ccb
--- /dev/null
+++ b/benchmark/car/gt.jsonl
@@ -0,0 +1,30 @@
+{"filename": "image/dm8WI-nJXmpN-kqg0C6gX.jpg", "image_height": 432, "image_width": 768, "instances": [{"label": 1, "label_name": "car", "bbox": [72, 286, 152, 353]}, {"label": 1, "label_name": "car", "bbox": [128, 272, 155, 292]}, {"label": 1, "label_name": "car", "bbox": [130, 247, 148, 263]}, {"label": 1, "label_name": "car", "bbox": [267, 267, 372, 351]}]}
+{"filename": "image/e8bCv7DetdVCCHV3-zRl3.jpg", "image_height": 800, "image_width": 1200, "instances": [{"label": 1, "label_name": "car", "bbox": [321, 78, 381, 199]}, {"label": 1, "label_name": "car", "bbox": [327, 222, 391, 340]}, {"label": 1, "label_name": "car", "bbox": [347, 357, 404, 472]}, {"label": 1, "label_name": "car", "bbox": [317, 489, 391, 612]}, {"label": 1, "label_name": "car", "bbox": [210, 609, 325, 675]}, {"label": 1, "label_name": "car", "bbox": [408, 408, 474, 529]}, {"label": 1, "label_name": "car", "bbox": [478, 348, 530, 477]}, {"label": 1, "label_name": "car", "bbox": [475, 78, 532, 209]}, {"label": 1, "label_name": "car", "bbox": [543, 121, 600, 245]}, {"label": 1, "label_name": "car", "bbox": [539, 9, 604, 85]}, {"label": 1, "label_name": "car", "bbox": [613, 136, 674, 270]}, {"label": 1, "label_name": "car", "bbox": [548, 270, 605, 384]}, {"label": 1, "label_name": "car", "bbox": [691, 283, 757, 416]}, {"label": 1, "label_name": "car", "bbox": [866, 98, 916, 211]}, {"label": 1, "label_name": "car", "bbox": [914, 13, 982, 95]}, {"label": 1, "label_name": "car", "bbox": [796, 402, 863, 532]}, {"label": 1, "label_name": "car", "bbox": [730, 503, 807, 621]}, {"label": 1, "label_name": "car", "bbox": [639, 515, 687, 617]}, {"label": 1, "label_name": "car", "bbox": [561, 484, 625, 600]}, {"label": 1, "label_name": "car", "bbox": [557, 610, 635, 728]}]}
+{"filename": "image/oXZx8CPpGIZzZl3nuj18A.jpg", "image_height": 496, "image_width": 999, "instances": [{"label": 1, "label_name": "car", "bbox": [449, 271, 546, 325]}]}
+{"filename": "image/n3i0yF2iBMpIP8RbzJg3A.jpg", "image_height": 799, "image_width": 1200, "instances": [{"label": 1, "label_name": "car", "bbox": [73, 206, 127, 242]}, {"label": 1, "label_name": "car", "bbox": [148, 191, 173, 216]}, {"label": 1, "label_name": "car", "bbox": [157, 154, 193, 184]}, {"label": 1, "label_name": "car", "bbox": [178, 159, 231, 186]}, {"label": 1, "label_name": "car", "bbox": [221, 158, 274, 189]}, {"label": 1, "label_name": "car", "bbox": [266, 164, 319, 187]}, {"label": 1, "label_name": "car", "bbox": [298, 160, 346, 194]}, {"label": 1, "label_name": "car", "bbox": [342, 162, 380, 195]}, {"label": 1, "label_name": "car", "bbox": [369, 166, 413, 198]}, {"label": 1, "label_name": "car", "bbox": [408, 176, 458, 210]}, {"label": 1, "label_name": "car", "bbox": [519, 195, 555, 226]}, {"label": 1, "label_name": "car", "bbox": [539, 179, 581, 204]}, {"label": 1, "label_name": "car", "bbox": [122, 213, 163, 252]}, {"label": 1, "label_name": "car", "bbox": [153, 213, 194, 251]}, {"label": 1, "label_name": "car", "bbox": [172, 194, 217, 221]}, {"label": 1, "label_name": "car", "bbox": [199, 196, 265, 221]}, {"label": 1, "label_name": "car", "bbox": [164, 214, 213, 259]}, {"label": 1, "label_name": "car", "bbox": [208, 224, 233, 258]}, {"label": 1, "label_name": "car", "bbox": [234, 219, 258, 258]}, {"label": 1, "label_name": "car", "bbox": [260, 223, 295, 266]}, {"label": 1, "label_name": "car", "bbox": [289, 218, 337, 256]}, {"label": 1, "label_name": "car", "bbox": [318, 216, 367, 265]}, {"label": 1, "label_name": "car", "bbox": [350, 200, 400, 225]}, {"label": 1, "label_name": "car", "bbox": [454, 218, 492, 248]}, {"label": 1, "label_name": "car", "bbox": [414, 201, 457, 232]}, {"label": 1, "label_name": "car", "bbox": [348, 224, 387, 280]}, {"label": 1, "label_name": "car", "bbox": [378, 222, 423, 267]}, {"label": 1, "label_name": "car", "bbox": [404, 232, 457, 274]}, {"label": 1, "label_name": "car", "bbox": [713, 176, 740, 207]}, {"label": 1, "label_name": "car", "bbox": [643, 143, 682, 167]}, {"label": 1, "label_name": "car", "bbox": [666, 105, 700, 127]}, {"label": 1, "label_name": "car", "bbox": [703, 108, 734, 137]}, {"label": 1, "label_name": "car", "bbox": [626, 104, 669, 130]}, {"label": 1, "label_name": "car", "bbox": [741, 171, 775, 206]}, {"label": 1, "label_name": "car", "bbox": [768, 168, 807, 213]}, {"label": 1, "label_name": "car", "bbox": [794, 179, 838, 220]}, {"label": 1, "label_name": "car", "bbox": [832, 184, 868, 217]}, {"label": 1, "label_name": "car", "bbox": [861, 185, 900, 220]}, {"label": 1, "label_name": "car", "bbox": [897, 187, 927, 218]}, {"label": 1, "label_name": "car", "bbox": [927, 188, 965, 227]}, {"label": 1, "label_name": "car", "bbox": [963, 189, 999, 224]}, {"label": 1, "label_name": "car", "bbox": [991, 195, 1032, 227]}, {"label": 1, "label_name": "car", "bbox": [1027, 197, 1067, 239]}, {"label": 1, "label_name": "car", "bbox": [1064, 200, 1099, 236]}, {"label": 1, "label_name": "car", "bbox": [1100, 198, 1141, 239]}, {"label": 1, "label_name": "car", "bbox": [1166, 255, 1198, 296]}, {"label": 1, "label_name": "car", "bbox": [1129, 252, 1165, 293]}, {"label": 1, "label_name": "car", "bbox": [1077, 249, 1122, 289]}, {"label": 1, "label_name": "car", "bbox": [1038, 244, 1080, 283]}, {"label": 1, "label_name": "car", "bbox": [1002, 242, 1035, 280]}, {"label": 1, "label_name": "car", "bbox": [967, 235, 1005, 276]}, {"label": 1, "label_name": "car", "bbox": [922, 236, 965, 269]}, {"label": 1, "label_name": "car", "bbox": [892, 234, 917, 268]}, {"label": 1, "label_name": "car", "bbox": [849, 231, 887, 261]}, {"label": 1, "label_name": "car", "bbox": [818, 226, 849, 259]}, {"label": 1, "label_name": "car", "bbox": [778, 226, 813, 255]}, {"label": 1, "label_name": "car", "bbox": [746, 222, 779, 253]}, {"label": 1, "label_name": "car", "bbox": [713, 215, 747, 246]}, {"label": 1, "label_name": "car", "bbox": [682, 216, 709, 243]}, {"label": 1, "label_name": "car", "bbox": [652, 233, 699, 268]}, {"label": 1, "label_name": "car", "bbox": [694, 233, 734, 275]}, {"label": 1, "label_name": "car", "bbox": [733, 243, 772, 288]}, {"label": 1, "label_name": "car", "bbox": [772, 243, 813, 291]}, {"label": 1, "label_name": "car", "bbox": [810, 244, 851, 295]}, {"label": 1, "label_name": "car", "bbox": [849, 258, 888, 309]}, {"label": 1, "label_name": "car", "bbox": [885, 268, 923, 307]}, {"label": 1, "label_name": "car", "bbox": [923, 266, 971, 307]}, {"label": 1, "label_name": "car", "bbox": [984, 259, 1026, 315]}, {"label": 1, "label_name": "car", "bbox": [1060, 270, 1109, 329]}, {"label": 1, "label_name": "car", "bbox": [1132, 279, 1183, 329]}, {"label": 1, "label_name": "car", "bbox": [666, 269, 738, 301]}, {"label": 1, "label_name": "car", "bbox": [610, 295, 646, 344]}, {"label": 1, "label_name": "car", "bbox": [640, 307, 693, 352]}, {"label": 1, "label_name": "car", "bbox": [677, 312, 740, 354]}, {"label": 1, "label_name": "car", "bbox": [776, 312, 819, 359]}, {"label": 1, "label_name": "car", "bbox": [827, 318, 872, 360]}, {"label": 1, "label_name": "car", "bbox": [869, 326, 925, 369]}, {"label": 1, "label_name": "car", "bbox": [921, 340, 965, 376]}, {"label": 1, "label_name": "car", "bbox": [968, 338, 1018, 387]}, {"label": 1, "label_name": "car", "bbox": [1017, 347, 1060, 392]}, {"label": 1, "label_name": "car", "bbox": [1062, 359, 1115, 403]}, {"label": 1, "label_name": "car", "bbox": [1112, 352, 1172, 397]}, {"label": 1, "label_name": "car", "bbox": [1168, 363, 1200, 414]}, {"label": 1, "label_name": "car", "bbox": [1152, 414, 1196, 490]}, {"label": 1, "label_name": "car", "bbox": [1086, 409, 1144, 486]}, {"label": 1, "label_name": "car", "bbox": [1032, 403, 1083, 481]}, {"label": 1, "label_name": "car", "bbox": [22, 257, 62, 286]}, {"label": 1, "label_name": "car", "bbox": [49, 254, 99, 296]}, {"label": 1, "label_name": "car", "bbox": [98, 261, 159, 295]}, {"label": 1, "label_name": "car", "bbox": [126, 266, 189, 301]}, {"label": 1, "label_name": "car", "bbox": [196, 266, 231, 305]}, {"label": 1, "label_name": "car", "bbox": [228, 277, 282, 300]}, {"label": 1, "label_name": "car", "bbox": [264, 282, 315, 308]}, {"label": 1, "label_name": "car", "bbox": [298, 285, 352, 308]}, {"label": 1, "label_name": "car", "bbox": [321, 287, 387, 323]}, {"label": 1, "label_name": "car", "bbox": [287, 311, 350, 364]}, {"label": 1, "label_name": "car", "bbox": [239, 313, 300, 364]}, {"label": 1, "label_name": "car", "bbox": [205, 305, 261, 356]}, {"label": 1, "label_name": "car", "bbox": [164, 303, 217, 356]}, {"label": 1, "label_name": "car", "bbox": [132, 298, 189, 347]}, {"label": 1, "label_name": "car", "bbox": [101, 294, 153, 345]}, {"label": 1, "label_name": "car", "bbox": [65, 293, 119, 342]}, {"label": 1, "label_name": "car", "bbox": [33, 295, 72, 340]}, {"label": 1, "label_name": "car", "bbox": [7, 377, 42, 420]}, {"label": 1, "label_name": "car", "bbox": [26, 377, 103, 422]}, {"label": 1, "label_name": "car", "bbox": [68, 389, 140, 425]}, {"label": 1, "label_name": "car", "bbox": [113, 388, 182, 426]}, {"label": 1, "label_name": "car", "bbox": [148, 389, 242, 450]}, {"label": 1, "label_name": "car", "bbox": [55, 444, 152, 516]}, {"label": 1, "label_name": "car", "bbox": [19, 430, 96, 509]}, {"label": 1, "label_name": "car", "bbox": [7, 422, 48, 499]}, {"label": 1, "label_name": "car", "bbox": [189, 454, 278, 517]}, {"label": 1, "label_name": "car", "bbox": [221, 472, 324, 553]}, {"label": 1, "label_name": "car", "bbox": [273, 495, 375, 576]}, {"label": 1, "label_name": "car", "bbox": [568, 336, 612, 392]}, {"label": 1, "label_name": "car", "bbox": [611, 350, 663, 415]}, {"label": 1, "label_name": "car", "bbox": [662, 351, 715, 421]}, {"label": 1, "label_name": "car", "bbox": [714, 353, 755, 420]}, {"label": 1, "label_name": "car", "bbox": [765, 360, 809, 428]}, {"label": 1, "label_name": "car", "bbox": [811, 365, 859, 443]}, {"label": 1, "label_name": "car", "bbox": [864, 375, 916, 454]}, {"label": 1, "label_name": "car", "bbox": [913, 381, 967, 457]}, {"label": 1, "label_name": "car", "bbox": [969, 395, 1021, 471]}, {"label": 1, "label_name": "car", "bbox": [4, 634, 91, 718]}, {"label": 1, "label_name": "car", "bbox": [481, 435, 546, 510]}, {"label": 1, "label_name": "car", "bbox": [524, 449, 609, 516]}, {"label": 1, "label_name": "car", "bbox": [585, 460, 656, 531]}, {"label": 1, "label_name": "car", "bbox": [647, 463, 724, 529]}, {"label": 1, "label_name": "car", "bbox": [714, 472, 780, 542]}, {"label": 1, "label_name": "car", "bbox": [775, 487, 837, 557]}, {"label": 1, "label_name": "car", "bbox": [832, 499, 909, 573]}, {"label": 1, "label_name": "car", "bbox": [910, 513, 972, 578]}, {"label": 1, "label_name": "car", "bbox": [982, 515, 1047, 607]}, {"label": 1, "label_name": "car", "bbox": [1042, 533, 1114, 621]}, {"label": 1, "label_name": "car", "bbox": [1116, 540, 1187, 623]}, {"label": 1, "label_name": "car", "bbox": [421, 503, 510, 597]}, {"label": 1, "label_name": "car", "bbox": [486, 511, 567, 603]}, {"label": 1, "label_name": "car", "bbox": [555, 529, 640, 627]}, {"label": 1, "label_name": "car", "bbox": [619, 532, 697, 636]}, {"label": 1, "label_name": "car", "bbox": [687, 526, 758, 656]}, {"label": 1, "label_name": "car", "bbox": [755, 548, 825, 669]}, {"label": 1, "label_name": "car", "bbox": [825, 573, 897, 685]}, {"label": 1, "label_name": "car", "bbox": [902, 573, 979, 695]}, {"label": 1, "label_name": "car", "bbox": [968, 607, 1056, 726]}, {"label": 1, "label_name": "car", "bbox": [1048, 632, 1135, 722]}, {"label": 1, "label_name": "car", "bbox": [1146, 629, 1198, 727]}]}
+{"filename": "image/vB4uib2oeDwoizQD2oE5T.jpg", "image_height": 708, "image_width": 1080, "instances": [{"label": 1, "label_name": "car", "bbox": [1, 281, 254, 453]}, {"label": 1, "label_name": "car", "bbox": [276, 282, 478, 385]}, {"label": 1, "label_name": "car", "bbox": [496, 257, 623, 343]}, {"label": 1, "label_name": "car", "bbox": [617, 246, 713, 320]}, {"label": 1, "label_name": "car", "bbox": [707, 255, 764, 301]}, {"label": 1, "label_name": "car", "bbox": [748, 247, 794, 290]}, {"label": 1, "label_name": "car", "bbox": [925, 214, 1001, 273]}]}
+{"filename": "image/ep1OYqiShiauTYnBt3jmV.jpg", "image_height": 966, "image_width": 1536, "instances": [{"label": 1, "label_name": "car", "bbox": [4, 607, 210, 882]}, {"label": 1, "label_name": "car", "bbox": [144, 664, 342, 798]}, {"label": 1, "label_name": "car", "bbox": [308, 618, 480, 720]}, {"label": 1, "label_name": "car", "bbox": [569, 545, 637, 605]}, {"label": 1, "label_name": "car", "bbox": [678, 594, 757, 623]}, {"label": 1, "label_name": "car", "bbox": [738, 564, 830, 598]}, {"label": 1, "label_name": "car", "bbox": [607, 619, 787, 732]}, {"label": 1, "label_name": "car", "bbox": [531, 730, 831, 876]}, {"label": 1, "label_name": "car", "bbox": [1208, 680, 1532, 880]}, {"label": 1, "label_name": "car", "bbox": [1099, 631, 1318, 829]}, {"label": 1, "label_name": "car", "bbox": [1034, 607, 1183, 718]}, {"label": 1, "label_name": "car", "bbox": [981, 596, 1071, 704]}, {"label": 1, "label_name": "car", "bbox": [940, 575, 1027, 662]}]}
+{"filename": "image/7QLIsS96jFQnvBid9KDvt.jpg", "image_height": 2276, "image_width": 1280, "instances": [{"label": 1, "label_name": "car", "bbox": [111, 1081, 193, 1147]}, {"label": 1, "label_name": "car", "bbox": [180, 1101, 265, 1163]}, {"label": 1, "label_name": "car", "bbox": [23, 1120, 131, 1182]}, {"label": 1, "label_name": "car", "bbox": [56, 1169, 167, 1231]}, {"label": 1, "label_name": "car", "bbox": [173, 1150, 271, 1212]}, {"label": 1, "label_name": "car", "bbox": [124, 1231, 213, 1297]}, {"label": 1, "label_name": "car", "bbox": [14, 1231, 109, 1293]}, {"label": 1, "label_name": "car", "bbox": [36, 1254, 115, 1336]}, {"label": 1, "label_name": "car", "bbox": [76, 1303, 184, 1369]}, {"label": 1, "label_name": "car", "bbox": [7, 1355, 96, 1417]}, {"label": 1, "label_name": "car", "bbox": [372, 830, 434, 922]}, {"label": 1, "label_name": "car", "bbox": [421, 850, 493, 919]}, {"label": 1, "label_name": "car", "bbox": [408, 684, 470, 756]}, {"label": 1, "label_name": "car", "bbox": [437, 677, 503, 746]}, {"label": 1, "label_name": "car", "bbox": [421, 589, 490, 678]}, {"label": 1, "label_name": "car", "bbox": [353, 1062, 445, 1124]}, {"label": 1, "label_name": "car", "bbox": [330, 1107, 432, 1169]}, {"label": 1, "label_name": "car", "bbox": [349, 1140, 454, 1212]}, {"label": 1, "label_name": "car", "bbox": [310, 1176, 408, 1245]}, {"label": 1, "label_name": "car", "bbox": [405, 1316, 497, 1385]}, {"label": 1, "label_name": "car", "bbox": [467, 1179, 565, 1241]}, {"label": 1, "label_name": "car", "bbox": [493, 1124, 588, 1186]}, {"label": 1, "label_name": "car", "bbox": [519, 1088, 637, 1157]}, {"label": 1, "label_name": "car", "bbox": [568, 1146, 686, 1221]}, {"label": 1, "label_name": "car", "bbox": [604, 1192, 725, 1261]}, {"label": 1, "label_name": "car", "bbox": [679, 1146, 797, 1208]}, {"label": 1, "label_name": "car", "bbox": [565, 1283, 689, 1352]}, {"label": 1, "label_name": "car", "bbox": [538, 1345, 653, 1417]}, {"label": 1, "label_name": "car", "bbox": [701, 1313, 812, 1392]}, {"label": 1, "label_name": "car", "bbox": [842, 1287, 934, 1349]}, {"label": 1, "label_name": "car", "bbox": [868, 1349, 979, 1421]}, {"label": 1, "label_name": "car", "bbox": [1106, 1349, 1201, 1421]}, {"label": 1, "label_name": "car", "bbox": [1174, 1303, 1263, 1365]}, {"label": 1, "label_name": "car", "bbox": [1018, 1208, 1136, 1277]}, {"label": 1, "label_name": "car", "bbox": [917, 1205, 1022, 1267]}, {"label": 1, "label_name": "car", "bbox": [939, 1130, 1034, 1192]}, {"label": 1, "label_name": "car", "bbox": [1168, 1163, 1260, 1225]}, {"label": 1, "label_name": "car", "bbox": [469, 508, 547, 574]}, {"label": 1, "label_name": "car", "bbox": [424, 451, 502, 521]}, {"label": 1, "label_name": "car", "bbox": [485, 435, 567, 513]}, {"label": 1, "label_name": "car", "bbox": [489, 370, 571, 448]}, {"label": 1, "label_name": "car", "bbox": [436, 288, 497, 362]}, {"label": 1, "label_name": "car", "bbox": [505, 89, 571, 155]}, {"label": 1, "label_name": "car", "bbox": [424, 110, 502, 180]}, {"label": 1, "label_name": "car", "bbox": [660, 179, 726, 253]}, {"label": 1, "label_name": "car", "bbox": [574, 329, 664, 415]}, {"label": 1, "label_name": "car", "bbox": [1005, 374, 1123, 435]}, {"label": 1, "label_name": "car", "bbox": [1095, 297, 1181, 367]}, {"label": 1, "label_name": "car", "bbox": [1009, 423, 1119, 505]}, {"label": 1, "label_name": "car", "bbox": [1017, 488, 1111, 558]}, {"label": 1, "label_name": "car", "bbox": [1038, 528, 1132, 610]}, {"label": 1, "label_name": "car", "bbox": [1172, 463, 1270, 529]}, {"label": 1, "label_name": "car", "bbox": [1172, 532, 1254, 638]}]}
+{"filename": "image/Ctud3AvEufQZRuk70byBK.jpg", "image_height": 259, "image_width": 390, "instances": [{"label": 1, "label_name": "car", "bbox": [122, 72, 156, 88]}, {"label": 1, "label_name": "car", "bbox": [208, 204, 245, 224]}]}
+{"filename": "image/A3dOllegcWLDe55BGf4iT.jpg", "image_height": 1153, "image_width": 1500, "instances": [{"label": 1, "label_name": "car", "bbox": [621, 630, 754, 735]}, {"label": 1, "label_name": "car", "bbox": [721, 802, 790, 905]}, {"label": 1, "label_name": "car", "bbox": [836, 812, 921, 915]}, {"label": 1, "label_name": "car", "bbox": [687, 940, 774, 1055]}, {"label": 1, "label_name": "car", "bbox": [830, 958, 925, 1073]}, {"label": 1, "label_name": "car", "bbox": [810, 1094, 909, 1143]}, {"label": 1, "label_name": "car", "bbox": [307, 238, 450, 443]}, {"label": 1, "label_name": "car", "bbox": [451, 356, 528, 461]}, {"label": 1, "label_name": "car", "bbox": [475, 298, 554, 357]}, {"label": 1, "label_name": "car", "bbox": [561, 352, 648, 427]}, {"label": 1, "label_name": "car", "bbox": [597, 280, 676, 353]}, {"label": 1, "label_name": "car", "bbox": [503, 216, 562, 289]}, {"label": 1, "label_name": "car", "bbox": [535, 164, 602, 231]}, {"label": 1, "label_name": "car", "bbox": [627, 194, 710, 267]}, {"label": 1, "label_name": "car", "bbox": [565, 96, 622, 159]}, {"label": 1, "label_name": "car", "bbox": [625, 140, 706, 199]}, {"label": 1, "label_name": "car", "bbox": [659, 24, 750, 121]}, {"label": 1, "label_name": "car", "bbox": [595, 34, 654, 99]}, {"label": 1, "label_name": "car", "bbox": [461, 6, 582, 105]}]}
+{"filename": "image/dtX90bO4KMb-EXSFAlLso.jpg", "image_height": 338, "image_width": 600, "instances": [{"label": 1, "label_name": "car", "bbox": [261, 151, 366, 235]}, {"label": 1, "label_name": "car", "bbox": [241, 111, 285, 130]}, {"label": 1, "label_name": "car", "bbox": [292, 108, 321, 124]}]}
+{"filename": "image/pGt8QoH9Z7AOTO_9YR7Ny.jpg", "image_height": 534, "image_width": 800, "instances": [{"label": 1, "label_name": "car", "bbox": [312, 399, 390, 456]}, {"label": 1, "label_name": "car", "bbox": [477, 397, 545, 456]}]}
+{"filename": "image/VjFL06hV9ge8Udp2KcyLh.jpg", "image_height": 534, "image_width": 800, "instances": [{"label": 1, "label_name": "car", "bbox": [214, 393, 251, 425]}, {"label": 1, "label_name": "car", "bbox": [323, 389, 350, 415]}, {"label": 1, "label_name": "car", "bbox": [502, 490, 594, 531]}, {"label": 1, "label_name": "car", "bbox": [471, 368, 496, 387]}]}
+{"filename": "image/RikwJaC9DJO2kXjm3naPE.jpg", "image_height": 450, "image_width": 800, "instances": [{"label": 1, "label_name": "car", "bbox": [78, 388, 221, 447]}, {"label": 1, "label_name": "car", "bbox": [222, 402, 248, 441]}, {"label": 1, "label_name": "car", "bbox": [148, 368, 206, 392]}, {"label": 1, "label_name": "car", "bbox": [167, 348, 219, 376]}, {"label": 1, "label_name": "car", "bbox": [223, 375, 276, 413]}, {"label": 1, "label_name": "car", "bbox": [235, 351, 280, 377]}, {"label": 1, "label_name": "car", "bbox": [314, 378, 369, 425]}, {"label": 1, "label_name": "car", "bbox": [326, 345, 367, 378]}, {"label": 1, "label_name": "car", "bbox": [254, 330, 302, 363]}, {"label": 1, "label_name": "car", "bbox": [223, 319, 260, 348]}, {"label": 1, "label_name": "car", "bbox": [244, 289, 294, 312]}, {"label": 1, "label_name": "car", "bbox": [221, 282, 250, 300]}, {"label": 1, "label_name": "car", "bbox": [333, 285, 395, 318]}, {"label": 1, "label_name": "car", "bbox": [388, 285, 433, 306]}, {"label": 1, "label_name": "car", "bbox": [446, 293, 506, 315]}, {"label": 1, "label_name": "car", "bbox": [514, 289, 573, 311]}, {"label": 1, "label_name": "car", "bbox": [364, 268, 387, 288]}]}
+{"filename": "image/AIXB3f3k2qtC0PxsZKqtc.jpg", "image_height": 525, "image_width": 678, "instances": [{"label": 1, "label_name": "car", "bbox": [339, 201, 411, 263]}, {"label": 1, "label_name": "car", "bbox": [524, 91, 623, 194]}]}
+{"filename": "image/Efki4ksB06bqavSbhhzyR.jpg", "image_height": 444, "image_width": 800, "instances": [{"label": 1, "label_name": "car", "bbox": [107, 284, 202, 384]}, {"label": 1, "label_name": "car", "bbox": [286, 270, 325, 310]}, {"label": 1, "label_name": "car", "bbox": [211, 400, 281, 442]}, {"label": 1, "label_name": "car", "bbox": [338, 289, 381, 328]}, {"label": 1, "label_name": "car", "bbox": [359, 245, 393, 270]}, {"label": 1, "label_name": "car", "bbox": [366, 225, 394, 245]}, {"label": 1, "label_name": "car", "bbox": [354, 215, 378, 235]}, {"label": 1, "label_name": "car", "bbox": [317, 232, 351, 252]}, {"label": 1, "label_name": "car", "bbox": [264, 198, 312, 260]}, {"label": 1, "label_name": "car", "bbox": [254, 186, 283, 224]}, {"label": 1, "label_name": "car", "bbox": [286, 168, 307, 194]}, {"label": 1, "label_name": "car", "bbox": [305, 174, 337, 212]}, {"label": 1, "label_name": "car", "bbox": [340, 161, 369, 192]}, {"label": 1, "label_name": "car", "bbox": [520, 164, 548, 197]}, {"label": 1, "label_name": "car", "bbox": [471, 185, 494, 199]}, {"label": 1, "label_name": "car", "bbox": [465, 174, 485, 189]}, {"label": 1, "label_name": "car", "bbox": [502, 185, 517, 202]}, {"label": 1, "label_name": "car", "bbox": [545, 195, 565, 213]}, {"label": 1, "label_name": "car", "bbox": [513, 198, 531, 219]}, {"label": 1, "label_name": "car", "bbox": [516, 211, 541, 230]}, {"label": 1, "label_name": "car", "bbox": [532, 225, 558, 247]}, {"label": 1, "label_name": "car", "bbox": [491, 219, 517, 240]}, {"label": 1, "label_name": "car", "bbox": [505, 237, 536, 264]}, {"label": 1, "label_name": "car", "bbox": [521, 271, 555, 303]}, {"label": 1, "label_name": "car", "bbox": [556, 246, 587, 273]}, {"label": 1, "label_name": "car", "bbox": [582, 270, 617, 294]}, {"label": 1, "label_name": "car", "bbox": [545, 306, 585, 338]}, {"label": 1, "label_name": "car", "bbox": [619, 309, 660, 344]}, {"label": 1, "label_name": "car", "bbox": [590, 347, 646, 388]}, {"label": 1, "label_name": "car", "bbox": [655, 298, 692, 330]}, {"label": 1, "label_name": "car", "bbox": [597, 214, 655, 278]}, {"label": 1, "label_name": "car", "bbox": [649, 209, 711, 283]}, {"label": 1, "label_name": "car", "bbox": [744, 307, 789, 340]}]}
+{"filename": "image/rLw9t28WNqJP5OnIrWVge.jpg", "image_height": 2160, "image_width": 3840, "instances": [{"label": 1, "label_name": "car", "bbox": [0, 401, 165, 510]}, {"label": 1, "label_name": "car", "bbox": [0, 540, 158, 649]}, {"label": 1, "label_name": "car", "bbox": [1020, 397, 1182, 495]}, {"label": 1, "label_name": "car", "bbox": [1357, 476, 1537, 570]}, {"label": 1, "label_name": "car", "bbox": [1871, 513, 2063, 615]}, {"label": 1, "label_name": "car", "bbox": [2448, 468, 2632, 562]}, {"label": 1, "label_name": "car", "bbox": [2715, 382, 2884, 487]}, {"label": 1, "label_name": "car", "bbox": [2966, 442, 3154, 525]}, {"label": 1, "label_name": "car", "bbox": [3015, 536, 3240, 608]}, {"label": 1, "label_name": "car", "bbox": [3656, 375, 3840, 462]}, {"label": 1, "label_name": "car", "bbox": [2808, 603, 3018, 690]}, {"label": 1, "label_name": "car", "bbox": [2370, 592, 2554, 675]}, {"label": 1, "label_name": "car", "bbox": [1691, 633, 1875, 716]}, {"label": 1, "label_name": "car", "bbox": [1342, 697, 1549, 776]}, {"label": 1, "label_name": "car", "bbox": [1102, 611, 1309, 701]}, {"label": 1, "label_name": "car", "bbox": [266, 1098, 465, 1192]}, {"label": 1, "label_name": "car", "bbox": [862, 1121, 1091, 1215]}, {"label": 1, "label_name": "car", "bbox": [971, 1350, 1174, 1452]}, {"label": 1, "label_name": "car", "bbox": [1406, 1143, 1601, 1245]}, {"label": 1, "label_name": "car", "bbox": [1766, 1263, 1973, 1372]}, {"label": 1, "label_name": "car", "bbox": [2246, 1065, 2453, 1155]}, {"label": 1, "label_name": "car", "bbox": [2253, 1151, 2467, 1279]}, {"label": 1, "label_name": "car", "bbox": [2617, 1383, 2831, 1473]}, {"label": 1, "label_name": "car", "bbox": [2827, 1256, 3052, 1369]}, {"label": 1, "label_name": "car", "bbox": [3082, 1083, 3277, 1170]}, {"label": 1, "label_name": "car", "bbox": [3243, 1173, 3453, 1282]}, {"label": 1, "label_name": "car", "bbox": [3045, 1908, 3300, 2047]}, {"label": 1, "label_name": "car", "bbox": [2077, 1905, 2340, 2022]}]}
+{"filename": "image/5CTI9lChqd6bsdlWMIGxE.jpg", "image_height": 626, "image_width": 418, "instances": [{"label": 1, "label_name": "car", "bbox": [184, 559, 264, 620]}, {"label": 1, "label_name": "car", "bbox": [186, 511, 232, 556]}, {"label": 1, "label_name": "car", "bbox": [188, 481, 229, 510]}, {"label": 1, "label_name": "car", "bbox": [145, 490, 177, 521]}, {"label": 1, "label_name": "car", "bbox": [98, 491, 135, 518]}, {"label": 1, "label_name": "car", "bbox": [156, 473, 185, 494]}, {"label": 1, "label_name": "car", "bbox": [163, 454, 187, 473]}, {"label": 1, "label_name": "car", "bbox": [195, 455, 222, 479]}]}
+{"filename": "image/6IJlRrwHnU2DNAHnhWbQI.jpg", "image_height": 949, "image_width": 1280, "instances": [{"label": 1, "label_name": "car", "bbox": [233, 607, 333, 666]}, {"label": 1, "label_name": "car", "bbox": [509, 574, 546, 617]}, {"label": 1, "label_name": "car", "bbox": [500, 556, 532, 583]}, {"label": 1, "label_name": "car", "bbox": [710, 536, 735, 564]}, {"label": 1, "label_name": "car", "bbox": [743, 533, 768, 563]}, {"label": 1, "label_name": "car", "bbox": [764, 548, 799, 580]}, {"label": 1, "label_name": "car", "bbox": [741, 574, 779, 609]}, {"label": 1, "label_name": "car", "bbox": [752, 656, 826, 737]}, {"label": 1, "label_name": "car", "bbox": [839, 623, 915, 687]}, {"label": 1, "label_name": "car", "bbox": [915, 762, 1112, 925]}]}
+{"filename": "image/E9TszYni5Gu1JDkoK-k2x.jpg", "image_height": 2160, "image_width": 3840, "instances": [{"label": 1, "label_name": "car", "bbox": [41, 1173, 146, 1233]}, {"label": 1, "label_name": "car", "bbox": [190, 1296, 295, 1352]}, {"label": 1, "label_name": "car", "bbox": [0, 744, 76, 804]}, {"label": 1, "label_name": "car", "bbox": [481, 964, 575, 1024]}, {"label": 1, "label_name": "car", "bbox": [597, 916, 698, 976]}, {"label": 1, "label_name": "car", "bbox": [857, 711, 951, 767]}, {"label": 1, "label_name": "car", "bbox": [1256, 722, 1361, 782]}, {"label": 1, "label_name": "car", "bbox": [943, 916, 1048, 972]}, {"label": 1, "label_name": "car", "bbox": [1491, 953, 1592, 1009]}, {"label": 1, "label_name": "car", "bbox": [1681, 1005, 1782, 1065]}, {"label": 1, "label_name": "car", "bbox": [1741, 908, 1838, 964]}, {"label": 1, "label_name": "car", "bbox": [1830, 841, 1927, 905]}, {"label": 1, "label_name": "car", "bbox": [2113, 886, 2240, 954]}, {"label": 1, "label_name": "car", "bbox": [2233, 949, 2338, 1017]}, {"label": 1, "label_name": "car", "bbox": [2348, 878, 2468, 938]}, {"label": 1, "label_name": "car", "bbox": [2527, 934, 2639, 994]}, {"label": 1, "label_name": "car", "bbox": [2799, 930, 2904, 986]}, {"label": 1, "label_name": "car", "bbox": [2803, 845, 2926, 909]}, {"label": 1, "label_name": "car", "bbox": [3064, 826, 3173, 890]}, {"label": 1, "label_name": "car", "bbox": [3216, 919, 3313, 990]}, {"label": 1, "label_name": "car", "bbox": [3328, 871, 3422, 927]}, {"label": 1, "label_name": "car", "bbox": [3574, 875, 3686, 931]}, {"label": 1, "label_name": "car", "bbox": [3604, 938, 3731, 998]}, {"label": 1, "label_name": "car", "bbox": [1577, 1202, 1693, 1258]}, {"label": 1, "label_name": "car", "bbox": [2590, 1296, 2687, 1364]}, {"label": 1, "label_name": "car", "bbox": [3269, 1639, 3363, 1740]}, {"label": 1, "label_name": "car", "bbox": [3604, 1486, 3660, 1595]}, {"label": 1, "label_name": "car", "bbox": [3656, 1478, 3720, 1594]}, {"label": 1, "label_name": "car", "bbox": [3731, 1482, 3795, 1583]}, {"label": 1, "label_name": "car", "bbox": [3358, 509, 3418, 618]}, {"label": 1, "label_name": "car", "bbox": [3421, 509, 3485, 603]}, {"label": 1, "label_name": "car", "bbox": [3485, 561, 3545, 666]}, {"label": 1, "label_name": "car", "bbox": [3406, 383, 3474, 488]}, {"label": 1, "label_name": "car", "bbox": [3406, 275, 3470, 369]}, {"label": 1, "label_name": "car", "bbox": [2337, 725, 2431, 781]}, {"label": 1, "label_name": "car", "bbox": [1163, 394, 1264, 454]}, {"label": 1, "label_name": "car", "bbox": [1320, 353, 1402, 409]}, {"label": 1, "label_name": "car", "bbox": [1431, 349, 1532, 405]}, {"label": 1, "label_name": "car", "bbox": [1562, 330, 1663, 401]}, {"label": 1, "label_name": "car", "bbox": [1677, 330, 1774, 405]}, {"label": 1, "label_name": "car", "bbox": [1763, 338, 1853, 402]}, {"label": 1, "label_name": "car", "bbox": [1014, 357, 1115, 413]}, {"label": 1, "label_name": "car", "bbox": [917, 345, 1011, 409]}, {"label": 1, "label_name": "car", "bbox": [790, 342, 880, 402]}, {"label": 1, "label_name": "car", "bbox": [701, 345, 795, 401]}, {"label": 1, "label_name": "car", "bbox": [693, 409, 798, 480]}, {"label": 1, "label_name": "car", "bbox": [276, 465, 340, 551]}, {"label": 1, "label_name": "car", "bbox": [198, 461, 254, 562]}, {"label": 1, "label_name": "car", "bbox": [1554, 450, 1610, 566]}, {"label": 1, "label_name": "car", "bbox": [1588, 439, 1644, 540]}, {"label": 1, "label_name": "car", "bbox": [2024, 431, 2092, 517]}, {"label": 1, "label_name": "car", "bbox": [2083, 424, 2151, 536]}, {"label": 1, "label_name": "car", "bbox": [2173, 424, 2233, 533]}, {"label": 1, "label_name": "car", "bbox": [2233, 439, 2293, 525]}, {"label": 1, "label_name": "car", "bbox": [2348, 323, 2442, 379]}, {"label": 1, "label_name": "car", "bbox": [2475, 316, 2580, 387]}]}
+{"filename": "image/5ZJ4NMRqk6apQnfLRvHAe.jpg", "image_height": 391, "image_width": 600, "instances": []}
+{"filename": "image/6GwqzUtcLIiiQVynsOsIe.jpg", "image_height": 1000, "image_width": 1500, "instances": [{"label": 1, "label_name": "car", "bbox": [539, 779, 645, 902]}, {"label": 1, "label_name": "car", "bbox": [592, 712, 682, 802]}, {"label": 1, "label_name": "car", "bbox": [377, 696, 484, 814]}, {"label": 1, "label_name": "car", "bbox": [1046, 691, 1153, 800]}, {"label": 1, "label_name": "car", "bbox": [1243, 769, 1363, 878]}, {"label": 1, "label_name": "car", "bbox": [1371, 896, 1489, 995]}, {"label": 1, "label_name": "car", "bbox": [1065, 610, 1174, 698]}, {"label": 1, "label_name": "car", "bbox": [1160, 600, 1271, 687]}, {"label": 1, "label_name": "car", "bbox": [896, 543, 1000, 645]}, {"label": 1, "label_name": "car", "bbox": [1000, 453, 1113, 531]}, {"label": 1, "label_name": "car", "bbox": [1088, 444, 1197, 529]}, {"label": 1, "label_name": "car", "bbox": [851, 396, 972, 505]}, {"label": 1, "label_name": "car", "bbox": [860, 320, 954, 391]}, {"label": 1, "label_name": "car", "bbox": [922, 288, 1019, 373]}, {"label": 1, "label_name": "car", "bbox": [705, 360, 795, 431]}, {"label": 1, "label_name": "car", "bbox": [644, 270, 717, 334]}, {"label": 1, "label_name": "car", "bbox": [686, 227, 766, 300]}, {"label": 1, "label_name": "car", "bbox": [767, 218, 849, 301]}, {"label": 1, "label_name": "car", "bbox": [798, 173, 883, 239]}, {"label": 1, "label_name": "car", "bbox": [888, 167, 983, 245]}, {"label": 1, "label_name": "car", "bbox": [762, 56, 844, 126]}, {"label": 1, "label_name": "car", "bbox": [731, 104, 814, 165]}, {"label": 1, "label_name": "car", "bbox": [639, 89, 733, 152]}, {"label": 1, "label_name": "car", "bbox": [665, 34, 745, 90]}, {"label": 1, "label_name": "car", "bbox": [532, 0, 610, 59]}, {"label": 1, "label_name": "car", "bbox": [453, 3, 521, 50]}, {"label": 1, "label_name": "car", "bbox": [360, 4, 436, 36]}, {"label": 1, "label_name": "car", "bbox": [539, 73, 622, 143]}, {"label": 1, "label_name": "car", "bbox": [603, 144, 679, 214]}, {"label": 1, "label_name": "car", "bbox": [505, 141, 588, 214]}, {"label": 1, "label_name": "car", "bbox": [1077, 94, 1159, 162]}, {"label": 1, "label_name": "car", "bbox": [1279, 268, 1371, 341]}, {"label": 1, "label_name": "car", "bbox": [408, 605, 502, 693]}, {"label": 1, "label_name": "car", "bbox": [494, 584, 588, 672]}, {"label": 1, "label_name": "car", "bbox": [667, 546, 774, 648]}, {"label": 1, "label_name": "car", "bbox": [582, 463, 688, 560]}, {"label": 1, "label_name": "car", "bbox": [522, 370, 617, 460]}, {"label": 1, "label_name": "car", "bbox": [454, 301, 541, 384]}, {"label": 1, "label_name": "car", "bbox": [303, 134, 386, 212]}, {"label": 1, "label_name": "car", "bbox": [44, 16, 120, 80]}, {"label": 1, "label_name": "car", "bbox": [87, 75, 172, 151]}, {"label": 1, "label_name": "car", "bbox": [140, 146, 230, 229]}, {"label": 1, "label_name": "car", "bbox": [230, 234, 313, 307]}, {"label": 1, "label_name": "car", "bbox": [135, 246, 223, 326]}, {"label": 1, "label_name": "car", "bbox": [73, 279, 161, 362]}, {"label": 1, "label_name": "car", "bbox": [272, 303, 364, 388]}, {"label": 1, "label_name": "car", "bbox": [192, 318, 270, 406]}, {"label": 1, "label_name": "car", "bbox": [154, 394, 241, 477]}, {"label": 1, "label_name": "car", "bbox": [256, 410, 338, 481]}, {"label": 1, "label_name": "car", "bbox": [353, 379, 436, 490]}, {"label": 1, "label_name": "car", "bbox": [216, 470, 299, 555]}, {"label": 1, "label_name": "car", "bbox": [318, 489, 408, 581]}, {"label": 1, "label_name": "car", "bbox": [406, 487, 503, 577]}, {"label": 1, "label_name": "car", "bbox": [275, 567, 376, 657]}]}
+{"filename": "image/rTxwtL7uOV0g57_k17y_K.jpeg", "image_height": 174, "image_width": 289, "instances": [{"label": 1, "label_name": "car", "bbox": [69, 94, 96, 119]}, {"label": 1, "label_name": "car", "bbox": [131, 114, 163, 140]}, {"label": 1, "label_name": "car", "bbox": [196, 102, 213, 118]}, {"label": 1, "label_name": "car", "bbox": [196, 95, 206, 102]}, {"label": 1, "label_name": "car", "bbox": [131, 93, 144, 103]}]}
+{"filename": "image/D3E6XeL439i-S95IpQOUn.jpeg", "image_height": 179, "image_width": 282, "instances": [{"label": 1, "label_name": "car", "bbox": [85, 64, 186, 109]}, {"label": 1, "label_name": "car", "bbox": [229, 69, 280, 115]}]}
+{"filename": "image/YxB2kc6-p6uvhKUFzqVFR.jpeg", "image_height": 200, "image_width": 252, "instances": [{"label": 1, "label_name": "car", "bbox": [129, 28, 142, 42]}, {"label": 1, "label_name": "car", "bbox": [138, 51, 152, 65]}, {"label": 1, "label_name": "car", "bbox": [142, 65, 158, 80]}, {"label": 1, "label_name": "car", "bbox": [150, 85, 167, 103]}, {"label": 1, "label_name": "car", "bbox": [159, 120, 179, 144]}, {"label": 1, "label_name": "car", "bbox": [168, 155, 194, 187]}, {"label": 1, "label_name": "car", "bbox": [183, 55, 198, 67]}, {"label": 1, "label_name": "car", "bbox": [191, 66, 209, 80]}, {"label": 1, "label_name": "car", "bbox": [208, 85, 227, 99]}, {"label": 1, "label_name": "car", "bbox": [218, 95, 249, 121]}, {"label": 1, "label_name": "car", "bbox": [174, 43, 183, 53]}, {"label": 1, "label_name": "car", "bbox": [164, 32, 174, 41]}]}
+{"filename": "image/fTQHmq6EYziz_JoUp1rnV.jpeg", "image_height": 178, "image_width": 282, "instances": [{"label": 1, "label_name": "car", "bbox": [51, 119, 97, 148]}, {"label": 1, "label_name": "car", "bbox": [90, 114, 103, 126]}, {"label": 1, "label_name": "car", "bbox": [108, 112, 121, 127]}, {"label": 1, "label_name": "car", "bbox": [0, 156, 20, 177]}, {"label": 1, "label_name": "car", "bbox": [146, 155, 168, 175]}, {"label": 1, "label_name": "car", "bbox": [172, 140, 189, 155]}, {"label": 1, "label_name": "car", "bbox": [189, 130, 202, 142]}, {"label": 1, "label_name": "car", "bbox": [239, 130, 280, 170]}, {"label": 1, "label_name": "car", "bbox": [203, 146, 228, 166]}, {"label": 1, "label_name": "car", "bbox": [151, 131, 163, 143]}, {"label": 1, "label_name": "car", "bbox": [144, 123, 152, 131]}, {"label": 1, "label_name": "car", "bbox": [167, 120, 177, 127]}, {"label": 1, "label_name": "car", "bbox": [171, 105, 180, 117]}, {"label": 1, "label_name": "car", "bbox": [155, 117, 162, 124]}]}
+{"filename": "image/kZ8pPPinTrxbL42ECZJ5J.jpg", "image_height": 595, "image_width": 1000, "instances": [{"label": 1, "label_name": "car", "bbox": [11, 319, 76, 368]}, {"label": 1, "label_name": "car", "bbox": [196, 271, 253, 309]}, {"label": 1, "label_name": "car", "bbox": [408, 302, 463, 338]}, {"label": 1, "label_name": "car", "bbox": [557, 291, 609, 329]}]}
+{"filename": "image/-E9TCQ2o9xNtxLppLLdgh.jpeg", "image_height": 300, "image_width": 533, "instances": [{"label": 1, "label_name": "car", "bbox": [1, 180, 54, 216]}, {"label": 1, "label_name": "car", "bbox": [0, 228, 30, 263]}, {"label": 1, "label_name": "car", "bbox": [6, 261, 72, 298]}, {"label": 1, "label_name": "car", "bbox": [95, 146, 147, 182]}, {"label": 1, "label_name": "car", "bbox": [99, 186, 156, 224]}, {"label": 1, "label_name": "car", "bbox": [192, 203, 248, 242]}, {"label": 1, "label_name": "car", "bbox": [227, 95, 280, 131]}, {"label": 1, "label_name": "car", "bbox": [243, 127, 296, 164]}, {"label": 1, "label_name": "car", "bbox": [307, 67, 358, 99]}, {"label": 1, "label_name": "car", "bbox": [320, 98, 371, 133]}, {"label": 1, "label_name": "car", "bbox": [384, 47, 421, 72]}, {"label": 1, "label_name": "car", "bbox": [392, 73, 441, 107]}, {"label": 1, "label_name": "car", "bbox": [450, 15, 503, 50]}, {"label": 1, "label_name": "car", "bbox": [450, 45, 506, 80]}, {"label": 1, "label_name": "car", "bbox": [507, 0, 532, 26]}, {"label": 1, "label_name": "car", "bbox": [512, 30, 531, 55]}]}
+{"filename": "image/I4lRcFpGml-MvNWer_J5j.jpg", "image_height": 500, "image_width": 750, "instances": [{"label": 1, "label_name": "car", "bbox": [106, 399, 145, 428]}, {"label": 1, "label_name": "car", "bbox": [181, 381, 224, 409]}, {"label": 1, "label_name": "car", "bbox": [235, 390, 271, 420]}, {"label": 1, "label_name": "car", "bbox": [247, 356, 275, 379]}, {"label": 1, "label_name": "car", "bbox": [293, 375, 321, 398]}, {"label": 1, "label_name": "car", "bbox": [318, 361, 337, 377]}, {"label": 1, "label_name": "car", "bbox": [324, 359, 373, 428]}, {"label": 1, "label_name": "car", "bbox": [304, 339, 320, 354]}, {"label": 1, "label_name": "car", "bbox": [355, 338, 372, 353]}, {"label": 1, "label_name": "car", "bbox": [408, 344, 425, 364]}, {"label": 1, "label_name": "car", "bbox": [429, 334, 443, 351]}, {"label": 1, "label_name": "car", "bbox": [468, 335, 484, 348]}, {"label": 1, "label_name": "car", "bbox": [468, 354, 494, 372]}, {"label": 1, "label_name": "car", "bbox": [444, 380, 474, 404]}, {"label": 1, "label_name": "car", "bbox": [490, 377, 522, 403]}, {"label": 1, "label_name": "car", "bbox": [456, 457, 513, 499]}, {"label": 1, "label_name": "car", "bbox": [568, 404, 619, 434]}, {"label": 1, "label_name": "car", "bbox": [406, 417, 445, 449]}, {"label": 1, "label_name": "car", "bbox": [155, 343, 179, 356]}]}
+{"filename": "image/OgD-MhEBsRxbwV2wsgZPE.jpg", "image_height": 500, "image_width": 889, "instances": [{"label": 1, "label_name": "car", "bbox": [364, 251, 413, 355]}, {"label": 1, "label_name": "car", "bbox": [406, 112, 439, 187]}]}
+{"filename": "image/u0CwaM61V6TeawWryt0DC.jpg", "image_height": 400, "image_width": 599, "instances": [{"label": 1, "label_name": "car", "bbox": [175, 220, 193, 242]}, {"label": 1, "label_name": "car", "bbox": [195, 217, 213, 235]}, {"label": 1, "label_name": "car", "bbox": [458, 252, 477, 279]}, {"label": 1, "label_name": "car", "bbox": [500, 287, 563, 327]}]}
diff --git a/benchmark/car/image/-E9TCQ2o9xNtxLppLLdgh.jpeg b/benchmark/car/image/-E9TCQ2o9xNtxLppLLdgh.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..1804e050efa2c9abee0157755026324b6029b93c
Binary files /dev/null and b/benchmark/car/image/-E9TCQ2o9xNtxLppLLdgh.jpeg differ
diff --git a/benchmark/car/image/5CTI9lChqd6bsdlWMIGxE.jpg b/benchmark/car/image/5CTI9lChqd6bsdlWMIGxE.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e87df00ebd082b4a38ed7887e2b58aeb2c334846
Binary files /dev/null and b/benchmark/car/image/5CTI9lChqd6bsdlWMIGxE.jpg differ
diff --git a/benchmark/car/image/5ZJ4NMRqk6apQnfLRvHAe.jpg b/benchmark/car/image/5ZJ4NMRqk6apQnfLRvHAe.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e570c533f47ff3b06d24c8db03436cbe060e9538
Binary files /dev/null and b/benchmark/car/image/5ZJ4NMRqk6apQnfLRvHAe.jpg differ
diff --git a/benchmark/car/image/6GwqzUtcLIiiQVynsOsIe.jpg b/benchmark/car/image/6GwqzUtcLIiiQVynsOsIe.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..02f29532397fc79f541d46e7aecb29bcb9fbb5a9
--- /dev/null
+++ b/benchmark/car/image/6GwqzUtcLIiiQVynsOsIe.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d5f47a82f942529a5781c0bad275c2e374a31f6552b87d88f09d46a1dbd2895
+size 710630
diff --git a/benchmark/car/image/6IJlRrwHnU2DNAHnhWbQI.jpg b/benchmark/car/image/6IJlRrwHnU2DNAHnhWbQI.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..08d7c543c3d4b04f978c737dd466a44459f0baa3
--- /dev/null
+++ b/benchmark/car/image/6IJlRrwHnU2DNAHnhWbQI.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e0713cae34a1bb1a2667476a300f96c5717cb248a742b2f6ea2c6b4b6e870da
+size 227068
diff --git a/benchmark/car/image/7QLIsS96jFQnvBid9KDvt.jpg b/benchmark/car/image/7QLIsS96jFQnvBid9KDvt.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..221fd9449659811f7183d6bdace51853d90f51d3
--- /dev/null
+++ b/benchmark/car/image/7QLIsS96jFQnvBid9KDvt.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ce20cb799c97591c0463539000c1bcff0321c84a0f95fd563755437faead0a2
+size 534896
diff --git a/benchmark/car/image/A3dOllegcWLDe55BGf4iT.jpg b/benchmark/car/image/A3dOllegcWLDe55BGf4iT.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8e99a7c6b8b80a620a3002049c3423184adc4317
--- /dev/null
+++ b/benchmark/car/image/A3dOllegcWLDe55BGf4iT.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51eb361e376bda95886645ffd515d2f69a13b89c399cb2729037954738a80702
+size 765787
diff --git a/benchmark/car/image/AIXB3f3k2qtC0PxsZKqtc.jpg b/benchmark/car/image/AIXB3f3k2qtC0PxsZKqtc.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ec798cc478877bc704c259053c10aed57dcb20c9
Binary files /dev/null and b/benchmark/car/image/AIXB3f3k2qtC0PxsZKqtc.jpg differ
diff --git a/benchmark/car/image/Ctud3AvEufQZRuk70byBK.jpg b/benchmark/car/image/Ctud3AvEufQZRuk70byBK.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..11b00516253e05f79c41afcb3f05723741033d44
Binary files /dev/null and b/benchmark/car/image/Ctud3AvEufQZRuk70byBK.jpg differ
diff --git a/benchmark/car/image/D3E6XeL439i-S95IpQOUn.jpeg b/benchmark/car/image/D3E6XeL439i-S95IpQOUn.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..2de525b25173b1b399976b29c925bd93462e66de
Binary files /dev/null and b/benchmark/car/image/D3E6XeL439i-S95IpQOUn.jpeg differ
diff --git a/benchmark/car/image/E9TszYni5Gu1JDkoK-k2x.jpg b/benchmark/car/image/E9TszYni5Gu1JDkoK-k2x.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8ec55fd1c3b5678217dbb32a69619cef9fe28811
--- /dev/null
+++ b/benchmark/car/image/E9TszYni5Gu1JDkoK-k2x.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52e1713c60dd511160901fcc2f0364235d82118cefaf567aeffbebc0dca18b37
+size 576822
diff --git a/benchmark/car/image/Efki4ksB06bqavSbhhzyR.jpg b/benchmark/car/image/Efki4ksB06bqavSbhhzyR.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..738027447bb85875384bd45e66c0c44365afc7bd
Binary files /dev/null and b/benchmark/car/image/Efki4ksB06bqavSbhhzyR.jpg differ
diff --git a/benchmark/car/image/I4lRcFpGml-MvNWer_J5j.jpg b/benchmark/car/image/I4lRcFpGml-MvNWer_J5j.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..99756bf1bc38d33a59116773c1b91c7a4b7f9cf3
Binary files /dev/null and b/benchmark/car/image/I4lRcFpGml-MvNWer_J5j.jpg differ
diff --git a/benchmark/car/image/OgD-MhEBsRxbwV2wsgZPE.jpg b/benchmark/car/image/OgD-MhEBsRxbwV2wsgZPE.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..753db40d6bdd92d27ee17ce69e3e08257127dc78
Binary files /dev/null and b/benchmark/car/image/OgD-MhEBsRxbwV2wsgZPE.jpg differ
diff --git a/benchmark/car/image/RikwJaC9DJO2kXjm3naPE.jpg b/benchmark/car/image/RikwJaC9DJO2kXjm3naPE.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..7d053cbc1604ba7cefdd80f16d37cba936454784
Binary files /dev/null and b/benchmark/car/image/RikwJaC9DJO2kXjm3naPE.jpg differ
diff --git a/benchmark/car/image/VjFL06hV9ge8Udp2KcyLh.jpg b/benchmark/car/image/VjFL06hV9ge8Udp2KcyLh.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3d57528adf72f00dd8bb233ee0c038dacab0d32e
Binary files /dev/null and b/benchmark/car/image/VjFL06hV9ge8Udp2KcyLh.jpg differ
diff --git a/benchmark/car/image/YxB2kc6-p6uvhKUFzqVFR.jpeg b/benchmark/car/image/YxB2kc6-p6uvhKUFzqVFR.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..2c26d8811f4c9d62031ff661d4fd5b44d14d99cb
Binary files /dev/null and b/benchmark/car/image/YxB2kc6-p6uvhKUFzqVFR.jpeg differ
diff --git a/benchmark/car/image/dm8WI-nJXmpN-kqg0C6gX.jpg b/benchmark/car/image/dm8WI-nJXmpN-kqg0C6gX.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ba2208b96c668fea62e8c24e57f0ccb80a1d38e6
--- /dev/null
+++ b/benchmark/car/image/dm8WI-nJXmpN-kqg0C6gX.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a3550ab9b19171a4ad73b6102c4d8bbeed9b953d480191e08d364f2937a453c
+size 180971
diff --git a/benchmark/car/image/dtX90bO4KMb-EXSFAlLso.jpg b/benchmark/car/image/dtX90bO4KMb-EXSFAlLso.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..1f1c53d28eea8a1aa98fc7ad847df3290e9e0e6e
Binary files /dev/null and b/benchmark/car/image/dtX90bO4KMb-EXSFAlLso.jpg differ
diff --git a/benchmark/car/image/e8bCv7DetdVCCHV3-zRl3.jpg b/benchmark/car/image/e8bCv7DetdVCCHV3-zRl3.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3d2d3f6f07891466f0c08febcfd01fe10978052b
--- /dev/null
+++ b/benchmark/car/image/e8bCv7DetdVCCHV3-zRl3.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc7b653cba4c9bd1582b15c95b60f53344614fa9515c94d266cd4b6d404cc023
+size 321090
diff --git a/benchmark/car/image/ep1OYqiShiauTYnBt3jmV.jpg b/benchmark/car/image/ep1OYqiShiauTYnBt3jmV.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..19adea7cde24cecefe32d7ac3a963c53b619fe88
--- /dev/null
+++ b/benchmark/car/image/ep1OYqiShiauTYnBt3jmV.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9dffde437011c131c06a1ad8ef026e2bfc1965499b7d4afeca88e6103c35814
+size 277648
diff --git a/benchmark/car/image/fTQHmq6EYziz_JoUp1rnV.jpeg b/benchmark/car/image/fTQHmq6EYziz_JoUp1rnV.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..a56c289f6b81d5e46cbeb8d8e7ee4f40adc54e7a
Binary files /dev/null and b/benchmark/car/image/fTQHmq6EYziz_JoUp1rnV.jpeg differ
diff --git a/benchmark/car/image/kZ8pPPinTrxbL42ECZJ5J.jpg b/benchmark/car/image/kZ8pPPinTrxbL42ECZJ5J.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4aa33246fcf3098fb078846b45a34fefec22d3b4
--- /dev/null
+++ b/benchmark/car/image/kZ8pPPinTrxbL42ECZJ5J.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecd16fd7f6cf1a5efc0e8ebdee289ace5de4dd19c55f2a67abaf035b58f286a8
+size 131353
diff --git a/benchmark/car/image/n3i0yF2iBMpIP8RbzJg3A.jpg b/benchmark/car/image/n3i0yF2iBMpIP8RbzJg3A.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..42b468e77aacd1a49c0409faaf2ea2d1c3291327
--- /dev/null
+++ b/benchmark/car/image/n3i0yF2iBMpIP8RbzJg3A.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a95a517408a9992908225db098433266c1a291bfd4dd97991ce19702752938b
+size 670137
diff --git a/benchmark/car/image/oXZx8CPpGIZzZl3nuj18A.jpg b/benchmark/car/image/oXZx8CPpGIZzZl3nuj18A.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..5ebe8926ba4e427862f3a76a48028e21bbf369f6
Binary files /dev/null and b/benchmark/car/image/oXZx8CPpGIZzZl3nuj18A.jpg differ
diff --git a/benchmark/car/image/pGt8QoH9Z7AOTO_9YR7Ny.jpg b/benchmark/car/image/pGt8QoH9Z7AOTO_9YR7Ny.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..24430462c86d426d1ede6cdf10f3a245939663d1
Binary files /dev/null and b/benchmark/car/image/pGt8QoH9Z7AOTO_9YR7Ny.jpg differ
diff --git a/benchmark/car/image/rLw9t28WNqJP5OnIrWVge.jpg b/benchmark/car/image/rLw9t28WNqJP5OnIrWVge.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..055b7f3dd5640242fc8942ca77d34996ad7485bc
--- /dev/null
+++ b/benchmark/car/image/rLw9t28WNqJP5OnIrWVge.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf4e9d88e469920994d1e393d3431e219bc870cba9aca7346018b080938a7385
+size 574838
diff --git a/benchmark/car/image/rTxwtL7uOV0g57_k17y_K.jpeg b/benchmark/car/image/rTxwtL7uOV0g57_k17y_K.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..02ea4099cac86f6c67511fc04cb29f8453858fc8
Binary files /dev/null and b/benchmark/car/image/rTxwtL7uOV0g57_k17y_K.jpeg differ
diff --git a/benchmark/car/image/u0CwaM61V6TeawWryt0DC.jpg b/benchmark/car/image/u0CwaM61V6TeawWryt0DC.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..d027354cdb5538df1e745419be9bffd379733327
--- /dev/null
+++ b/benchmark/car/image/u0CwaM61V6TeawWryt0DC.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a66ee64998ce503265897c3af20bd1917ab3abced189638f9086ead7e17a78a
+size 397138
diff --git a/benchmark/car/image/vB4uib2oeDwoizQD2oE5T.jpg b/benchmark/car/image/vB4uib2oeDwoizQD2oE5T.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3af876923260423c445f42ecaa381b9355d99b9f
--- /dev/null
+++ b/benchmark/car/image/vB4uib2oeDwoizQD2oE5T.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b2e42641df91d8644486583a9298e9801d39bd6ffc7383becbde260998a0554
+size 218837
diff --git a/benchmark/car/image_description.jsonl b/benchmark/car/image_description.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..9b9cde2ac53842263a6f9491ed1a3d755473dcec
--- /dev/null
+++ b/benchmark/car/image_description.jsonl
@@ -0,0 +1,29 @@
+{"image_path": "image/dm8WI-nJXmpN-kqg0C6gX.jpg", "image_description": "The image shows a toll booth area on a highway in Thailand. There are two lanes for payment, with one lane labeled \"เงินสด/เติมเงิน\" (Cash/Top Up) and another labeled \"อัตโนมัติ\" (Easy Pass). The Easy Pass lane has a sign indicating it accepts payments from 4 lanes. There are traffic lights above each lane, and the road appears to be clear with a few vehicles passing through. The photo also includes a watermark at the bottom right corner that reads \"www.thaizhonghua.com.\""}
+{"image_path": "image/e8bCv7DetdVCCHV3-zRl3.jpg", "image_description": "The image shows an aerial view of a busy intersection at night. There are multiple lanes of traffic with cars moving in different directions. The road markings are clearly visible, including pedestrian crossings and lane dividers. On both sides of the intersection, there are groups of people gathered, possibly waiting to cross or simply standing around. The vehicles vary in color and model, with some appearing to be taxis due to their distinctive green color.\nRegarding the target car, it is not specified which particular car you are referring to. If you have a specific car in mind, please provide more details such as its color, make, model, or any other distinguishing features, and I can help identify it for you."}
+{"image_path": "image/oXZx8CPpGIZzZl3nuj18A.jpg", "image_description": "The image shows an aerial view of a street intersection at night. The road markings include yellow and white lines, with a pedestrian crossing visible on the left side. There is a black car positioned near the center of the image, slightly angled towards the bottom right corner. The car appears to be stationary or moving very slowly. The surrounding area includes some greenery and a sidewalk with a few people standing near the edge. The lighting suggests it is nighttime, and there are no other vehicles immediately around the target car."}
+{"image_path": "image/n3i0yF2iBMpIP8RbzJg3A.jpg", "image_description": "The image depicts an aerial view of a parking lot at night. The parking lot is well-lit and contains numerous cars parked in designated spaces. The cars vary in color and model, with some appearing to be newer models while others look older. The parking lot is divided into sections, each with its own set of parking spaces. There is a covered walkway or tunnel that runs through the center of the parking lot, providing shelter for pedestrians. In the background, there are buildings with illuminated windows, indicating that it is evening or nighttime. The overall scene suggests a busy urban area with a high demand for parking."}
+{"image_path": "image/vB4uib2oeDwoizQD2oE5T.jpg", "image_description": "The image shows a nighttime urban street scene with several vehicles and traffic cones. The road has a green-painted lane on the right side, which appears to be designated for a specific purpose, possibly a bike lane or a bus lane, as indicated by the white markings and the presence of traffic cones. There are orange traffic cones placed along the edge of this green lane, likely to guide traffic or mark the boundary.\nIn the foreground, there is a silver minivan driving on the left side of the road. Behind it, there is a yellow taxi and another vehicle that appears to be a black SUV. The background features a row of shops with illuminated signs, including brands like Puma and New Balance. The streetlights and shop signs are brightly lit, creating a vibrant nighttime atmosphere. The overall scene suggests a busy commercial area."}
+{"image_path": "image/ep1OYqiShiauTYnBt3jmV.jpg", "image_description": "The image depicts a busy urban scene with a large crowd gathered on both sides of a road and on an elevated walkway. The road is lined with police vehicles, indicating some form of organized event or protest. There are several cars visible, including a black Audi SUV in the foreground. The background features modern buildings, a clock tower, and a cloudy sky. The overall atmosphere suggests a significant public gathering, possibly related to a demonstration or public event."}
+{"image_path": "image/7QLIsS96jFQnvBid9KDvt.jpg", "image_description": "The image shows an aerial view of a busy urban intersection with multiple lanes of traffic. The roads are divided into several lanes, and there are pedestrian crosswalks visible at the intersection. The area is surrounded by buildings of varying heights, some of which appear to be residential or commercial structures. There are trees planted along the sidewalks and in landscaped areas near the road.\nIn terms of identifying a specific target car, the image contains numerous cars, but without additional context or a clear indication of what constitutes the \"target,\" it's not possible to identify a single car as the target. If you have a specific car in mind that you would like identified, please provide more details or a reference point within the image."}
+{"image_path": "image/Ctud3AvEufQZRuk70byBK.jpg", "image_description": "The image shows an aerial view of a multi-lane highway with two cars visible on the road. The highway has multiple lanes in each direction, separated by a median strip with greenery and streetlights. The cars appear to be moving at a moderate speed. The target car is located in the lower part of the image, closer to the bottom edge, and it is a dark-colored vehicle, possibly black or dark gray. It is positioned in the rightmost lane of the lower section of the highway."}
+{"image_path": "image/A3dOllegcWLDe55BGf4iT.jpg", "image_description": "The image shows an aerial view of a busy urban intersection with multiple lanes of traffic. There are several vehicles, including cars and buses, moving through the intersection. The road markings include directional arrows indicating the flow of traffic and pedestrian crossing lines. The surrounding area includes streetlights, trees, and a paved plaza to the right.\nRegarding the target car, it appears to be a white SUV positioned near the center of the intersection, slightly angled towards the bottom left corner of the image. It is stopped at what seems to be a red light, as indicated by the traffic signal above it. The vehicle has a modern design, with visible headlights and taillights. There are no other specific details about the make or model of the car from this angle."}
+{"image_path": "image/dtX90bO4KMb-EXSFAlLso.jpg", "image_description": "The image appears to be a security camera capture taken on January 22, 2023, at 14:11:36. The scene is set in an urban area with wet roads, suggesting recent rain. There are several storefronts and signs visible, including one that reads \"心缘店\" (Heart缘 Store) and another that says \"华服饰\" (Hua Clothing). A small pickup truck is driving down the street from left to right. On the right side of the image, there is a person wearing a red jacket standing near a doorway or entrance. The environment looks somewhat busy but not crowded. The timestamp indicates the location's coordinates as 32.39.76310N 120.25.31630E."}
+{"image_path": "image/pGt8QoH9Z7AOTO_9YR7Ny.jpg", "image_description": "The image depicts a scenic road winding through a lush, hilly landscape. The road is bordered by stone walls and metal railings on both sides, with trees and vegetation lining the path. In the background, there are houses nestled among the hills, suggesting a residential area. The sky is clear and blue, indicating a sunny day.\nThere are two cars visible in the image. The first car is a white SUV driving away from the camera, positioned closer to the left side of the road. The second car is a black van also driving away from the camera, located further down the road towards the right side. Both vehicles appear to be moving at a moderate speed. The road itself is paved and appears to be well-maintained."}
+{"image_path": "image/VjFL06hV9ge8Udp2KcyLh.jpg", "image_description": "The image depicts a wide, multi-lane highway under a bright blue sky with scattered clouds. The road is flanked by trees and greenery on both sides, with some purple flowering trees adding a splash of color. In the background, there is a tall structure that resembles a telecommunications tower. The scene appears to be taken during the daytime, given the sunlight and clear visibility.\nThere are several cars visible on the road, including a white car in the foreground on the right side of the image, a yellow car further ahead, and other vehicles in the distance. The road has multiple lanes, separated by barriers, and the overall atmosphere suggests a busy yet orderly urban environment."}
+{"image_path": "image/RikwJaC9DJO2kXjm3naPE.jpg", "image_description": "The image depicts a busy urban street at night, illuminated by various lights from vehicles and street lamps. The scene includes multiple lanes of traffic with cars, motorcycles, and buses moving in both directions. Overhead, there is an elevated train track with a train passing through it. The architecture in the background features modern buildings with glass facades and some traditional structures. Trees line the sides of the road, adding greenery to the urban environment.\nRegarding the target car, the image does not specify any particular car as a target. However, if you are referring to a specific car in the image, please provide more details or a bounding box around the car you are interested in, and I can describe its characteristics based on that information."}
+{"image_path": "image/AIXB3f3k2qtC0PxsZKqtc.jpg", "image_description": "The image appears to be a security camera capture from a road surveillance system. The timestamp on the image indicates it was taken at 17:44 on October 6, 2022. The location is identified as \"甬金高速G1512 K35+200 中央-两头坞隧道广场 [奉化]\" which translates to \"Nongjin Expressway G1512 K35+200 Central - Liangtouwudu Tunnel Square [Fenghua]\". The camera number is \"CAM-48\".\nIn the image, there is a two-lane road with a central barrier. A black SUV is driving towards the camera on the left lane, and a large blue truck is driving away from the camera on the right lane. The road is surrounded by greenery, including trees and bushes, and there is a building visible in the background on the left side of the image. The road appears to be slightly curved, and there is a pedestrian crossing painted on the road. The overall lighting suggests it might be evening or late afternoon."}
+{"image_path": "image/Efki4ksB06bqavSbhhzyR.jpg", "image_description": "The image shows a busy bridge over a body of water, likely a river or a wide canal, with multiple lanes of traffic moving in both directions. The bridge has a modern design with a steel framework and is supported by large blue trusses. There are numerous vehicles on the bridge, including cars, trucks, and buses, indicating heavy traffic flow. The weather appears to be clear, and the sky is slightly hazy.\nIn terms of identifying a specific target car, the image does not have any particular car marked as a target. However, if you point out a specific vehicle you're interested in, I can provide more details about it."}
+{"image_path": "image/rLw9t28WNqJP5OnIrWVge.jpg", "image_description": "The image shows an aerial view of a multi-lane highway with traffic moving in both directions. The road is flanked by greenery, including trees and landscaped areas with wavy patterns of shrubs and flowers. There are several cars visible on the road, with some in motion and others appearing to be stationary or turning.\nThe target car is located in the lower right section of the image, traveling in the direction indicated by the white arrow painted on the road. It is a white vehicle, possibly a sedan, and it appears to be in the process of turning onto a side street or ramp. The car is positioned near the edge of the main highway, close to the landscaped area."}
+{"image_path": "image/5CTI9lChqd6bsdlWMIGxE.jpg", "image_description": "The image depicts an urban street scene at night. The sky is dark, and the area is illuminated by the lights from tall buildings and streetlights. There are multiple lanes of traffic with cars moving in both directions. The buildings in the background are modern and have numerous lit windows, suggesting it might be late evening or nighttime.\nRegarding the target car, there isn't a specific \"target\" car mentioned in your request. However, if you're referring to a particular car in the image, I can describe it based on its position and appearance. For example, if you point out a specific car, I can describe its color, size, and any distinctive features. If you need help identifying a specific car, please indicate which one you're interested in."}
+{"image_path": "image/6IJlRrwHnU2DNAHnhWbQI.jpg", "image_description": "The image depicts a city street at dusk or early evening. The sky is a gradient of blue hues, transitioning from a darker shade at the top to a lighter one near the horizon. The street is well-lit with streetlights, and there are multiple lanes of traffic moving in both directions. The road markings are clearly visible, including lane dividers and directional arrows.\nThere are several cars on the road, with their headlights and taillights illuminated. The vehicles vary in color and model, but specific details about individual cars cannot be discerned due to the distance and lighting conditions. The buildings on either side of the street are modern, with some featuring glass facades. There are also trees lining the sidewalk, which appear to be bare, suggesting it might be late autumn or winter.\nThe overall atmosphere of the image is calm and orderly, typical of a busy urban environment during the evening hours."}
+{"image_path": "image/E9TszYni5Gu1JDkoK-k2x.jpg", "image_description": "The image is an aerial view of a multi-lane urban road intersection. The road is divided into multiple lanes with clear lane markings and directional arrows indicating traffic flow. There are several cars visible on the road, moving in different directions. On the left side of the image, there is a row of parked cars along the curb, and some greenery is visible between the road and the parking area. The surrounding area includes buildings and what appears to be a mix of residential and commercial structures. The overall scene suggests a typical city environment with organized traffic and infrastructure.\nRegarding the target car, it is not specified which particular car you are referring to. If you have a specific car in mind, please provide more details or a reference point, and I can assist you further."}
+{"image_path": "image/6GwqzUtcLIiiQVynsOsIe.jpg", "image_description": "The image shows an aerial view of a busy urban intersection with multiple lanes of traffic. There are several cars and a few motorcycles visible on the roads. The vehicles appear to be moving in both directions, with some cars stopped at a traffic light. The road is divided into lanes with clear markings, including arrows indicating the direction of traffic flow. There are also pedestrian crosswalks visible, and greenery is planted along the sides of the road. A person can be seen walking near the pedestrian crossing.\nRegarding the target car, there isn't a specific car highlighted or marked as the \"target\" in the image. If you have a particular car in mind that you would like me to describe, please provide more details or specify which car you are referring to."}
+{"image_path": "image/rTxwtL7uOV0g57_k17y_K.jpeg", "image_description": "The image depicts a multi-lane highway with several vehicles traveling on it. The road is surrounded by greenery and hills in the background, suggesting a rural or semi-rural setting. There are signs overhead indicating directions and possibly speed limits.\nThe target car is a silver sedan located in the center lane of the highway. It appears to be moving at a moderate speed, following the flow of traffic. The car's position relative to other vehicles suggests it is maintaining a safe distance from those around it. The license plate of the car is visible but not legible in the image."}
+{"image_path": "image/D3E6XeL439i-S95IpQOUn.jpeg", "image_description": "The image shows a scene where a car has flipped over on its side on a road. There is smoke rising from the area where the car is located, indicating that there might be fire or some other form of damage. Several people are gathered around the overturned vehicle, appearing to assess the situation or possibly attempting to help. The background includes trees and what looks like a residential area with houses. The target car is the one that is upside down, with its wheels in the air and the body of the car facing upwards. It appears to be a white car, and it is clearly the focus of the scene due to its unusual position."}
+{"image_path": "image/YxB2kc6-p6uvhKUFzqVFR.jpeg", "image_description": "The image depicts a nighttime scene of a busy street market or fair. The area is illuminated by numerous lights from stalls and vehicles, creating a vibrant atmosphere. There are several rows of stalls with colorful canopies, likely selling various goods such as food, crafts, and souvenirs. The street is lined with parked cars, indicating that it is a popular spot for visitors. The road appears to be a two-way street with some traffic flowing in both directions.\nRegarding the target car, there isn't a specific car highlighted or marked in the image. If you have a particular car in mind that you would like me to describe, please provide more details or specify which car you are referring to."}
+{"image_path": "image/fTQHmq6EYziz_JoUp1rnV.jpeg", "image_description": "The image depicts a busy urban street scene with multiple lanes of traffic and tall buildings on both sides. The road is lined with trees and there are several vehicles visible, including buses and cars. The sky is clear and blue, suggesting it is a sunny day.\nThe target car appears to be a red sedan located in the right lane, slightly ahead of a black sedan. It is positioned near the center of the image, moving forward along the road. The car's color contrasts with the surrounding vehicles and the greenery of the trees."}
+{"image_path": "image/kZ8pPPinTrxbL42ECZJ5J.jpg", "image_description": "The image depicts a multi-lane road intersection with traffic lights and vehicles. The road has multiple lanes in each direction, separated by a median strip. There are several cars visible on the road, including a white car in the foreground and a dark-colored car further back. The background features trees, a palm tree, and mountains under a clear blue sky. A speed limit sign indicating \"15\" is visible on the right side of the image. The overall scene appears to be in a suburban or semi-urban area."}
+{"image_path": "image/-E9TCQ2o9xNtxLppLLdgh.jpeg", "image_description": "The image shows an aerial view of a busy urban road with multiple lanes of traffic. The road is lined with trees on both sides, and there are sidewalks visible as well. Several cars are seen traveling in both directions, with some vehicles stopped or moving slowly. The road appears to be part of a larger city infrastructure, with clear lane markings and traffic signals.\nThere is no specific \"target car\" highlighted in the image. If you're referring to a particular vehicle, please provide more details such as its color, model, or position within the scene for a more accurate description."}
+{"image_path": "image/I4lRcFpGml-MvNWer_J5j.jpg", "image_description": "The image depicts a bustling urban scene with multiple lanes of traffic on a wide road. The road is flanked by tall buildings, some of which are under construction, as indicated by the crane visible in the background. The sky is clear and blue, suggesting a sunny day. There are numerous cars and trucks moving along the road, and the area appears to be well-maintained with greenery lining the sides of the road.\nRegarding the target car, there isn't a specific car highlighted or marked in the image. If you have a particular car in mind that you would like me to describe, please provide more details or specify which part of the image you are referring to."}
+{"image_path": "image/OgD-MhEBsRxbwV2wsgZPE.jpg", "image_description": "The image shows an aerial view of a road with multiple lanes. There are two vehicles visible on the road: a white car and a larger vehicle, possibly a truck or bus, which appears to be a silver or gray color. The road is flanked by green trees on both sides, and there are streetlights along the road. The road has clear lane markings, including dashed lines indicating the center and solid lines for the edges of the lanes. The overall scene suggests a calm, possibly suburban or semi-urban area."}
+{"image_path": "image/u0CwaM61V6TeawWryt0DC.jpg", "image_description": "The image depicts a busy urban street at night, illuminated by streetlights and various neon signs. The scene is bustling with activity, featuring numerous motorcycles and scooters moving along the road, as well as some cars. The road appears to be a multi-lane highway with a pedestrian crossing visible in the foreground. On the right side, there's a large billboard advertising \"Kangaroo Hydrogen.\" The background showcases tall buildings with lit windows, suggesting a cityscape. The overall atmosphere is lively and dynamic, indicative of a metropolitan area during peak hours.\nRegarding the target car: There isn't a specific target car highlighted or marked in the image. If you're referring to a particular vehicle, please clarify which one you'd like me to describe."}
diff --git a/benchmark/car/user_input.txt b/benchmark/car/user_input.txt
new file mode 100644
index 0000000000000000000000000000000000000000..106e0b33a15f6e9da66adad54b20888081cbb945
--- /dev/null
+++ b/benchmark/car/user_input.txt
@@ -0,0 +1 @@
+检测道路上的车辆
\ No newline at end of file
diff --git "a/benchmark/car/\345\217\202\350\200\203\345\233\276\347\211\207/-E9TCQ2o9xNtxLppLLdgh.jpeg" "b/benchmark/car/\345\217\202\350\200\203\345\233\276\347\211\207/-E9TCQ2o9xNtxLppLLdgh.jpeg"
new file mode 100644
index 0000000000000000000000000000000000000000..1804e050efa2c9abee0157755026324b6029b93c
Binary files /dev/null and "b/benchmark/car/\345\217\202\350\200\203\345\233\276\347\211\207/-E9TCQ2o9xNtxLppLLdgh.jpeg" differ
diff --git a/benchmark/car_plate/gt.jsonl b/benchmark/car_plate/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..594c45f66faf74480de68a0677042a05468fe839
--- /dev/null
+++ b/benchmark/car_plate/gt.jsonl
@@ -0,0 +1,30 @@
+{"filename": "image/#9170169492945964687.jpg", "image_height": 0, "image_width": 0, "instances": [{"bbox": [269.6, 130.0, 351.6, 167.6], "is_ignored": false, "label": 1}]}
+{"filename": "image/4966998296728060233.jpg", "image_height": 0, "image_width": 0, "instances": [{"bbox": [311.7, 137.8, 432.5, 194.9], "is_ignored": false, "label": 1}]}
+{"filename": "image/#2563290238964912476.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [536.46, 665.87, 598.76, 700.98]}]}
+{"filename": "image/1671412636281891010.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [182.3, 317.8, 255.5, 337.8]}]}
+{"filename": "image/3021668729827686673.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [715.68, 730.27, 865.19, 777.19]}]}
+{"filename": "image/#3334113938196704587.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [260.71, 425.35, 462.71, 571.51]}]}
+{"filename": "image/#6659591284551536526.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [131.55, 254.28, 178.87, 280.2]}]}
+{"filename": "image/1794121684277643246.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [514.89, 770.37, 646.59, 811.71]}]}
+{"filename": "image/#1500244263428553553.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [325.43, 421.6, 477.07, 534.42]}]}
+{"filename": "image/#8813093836125699391.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [167.01, 294.59, 259.72, 324.21]}]}
+{"filename": "image/#2249067055978906958.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [176.1, 428.55, 468.93, 638.93]}]}
+{"filename": "image/582186526753952006.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [235.99, 337.2, 384.4, 384.14]}]}
+{"filename": "image/#3517741965209098503.jpg", "image_height": 0, "image_width": 0, "instances": [{"bbox": [315.5, 152.1, 421.4, 200.4], "is_ignored": false, "label": 1}]}
+{"filename": "image/6386760609565435452.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [243.34, 572.48, 358.94, 611.12]}]}
+{"filename": "image/318568126668983171.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [1018.65, 2656.58, 1265.64, 2759.97]}]}
+{"filename": "image/5321180974886966504.jpg", "image_height": 0, "image_width": 0, "instances": [{"bbox": [324.9, 102.3, 396.7, 139.5], "is_ignored": false, "label": 1}]}
+{"filename": "image/#7381771666275058687.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [533.48, 568.8, 715.77, 626.45]}]}
+{"filename": "image/#7518111569611972375.jpg", "image_height": 0, "image_width": 0, "instances": [{"bbox": [362.7, 95.3, 464.5, 141.6], "is_ignored": false, "label": 1}]}
+{"filename": "image/1140623623523992149.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [201.62, 375.73, 461.18, 510.21]}]}
+{"filename": "image/2579392287500286667.jpg", "image_height": 0, "image_width": 0, "instances": [{"bbox": [160.0, 262.4, 282.6, 315.9], "is_ignored": false, "label": 1}]}
+{"filename": "image/#6238260831625012504.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [115.49, 217.16, 204.47, 242.62]}]}
+{"filename": "image/#3630149080300033894.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [199.61, 357.6, 290.69, 385.39]}]}
+{"filename": "image/1850854682574030615.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [431.51, 487.35, 586.29, 625.58]}]}
+{"filename": "image/#7134505881093256854.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [310.17, 492.68, 530.01, 657.29]}]}
+{"filename": "image/#761631743535467855.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [213.87, 452.99, 483.57, 598.99]}]}
+{"filename": "image/#8529343536891984474.jpg", "image_height": 0, "image_width": 0, "instances": [{"bbox": [282.2, 144.6, 385.0, 195.3], "is_ignored": false, "label": 1}]}
+{"filename": "image/3488762191409319384.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [463.88, 489.11, 612.56, 535.31]}]}
+{"filename": "image/438344865304307623.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [120.83, 165.81, 249.21, 205.87]}]}
+{"filename": "image/#6227411411065045336.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [450.35, 908.47, 640.98, 961.75]}]}
+{"filename": "image/#286741634995118034.jpg", "image_height": 0, "image_width": 0, "instances": [{"is_ignored": false, "label": 1, "bbox": [253.05, 473.77, 395.75, 508.45]}]}
diff --git a/benchmark/car_plate/image/#1500244263428553553.jpg b/benchmark/car_plate/image/#1500244263428553553.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f7a99dee27f0f18ece5d18188eb93d19e73bac88
Binary files /dev/null and b/benchmark/car_plate/image/#1500244263428553553.jpg differ
diff --git a/benchmark/car_plate/image/#2249067055978906958.jpg b/benchmark/car_plate/image/#2249067055978906958.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e27c6a69d0fcc734b95188e03bfda4a33726269d
Binary files /dev/null and b/benchmark/car_plate/image/#2249067055978906958.jpg differ
diff --git a/benchmark/car_plate/image/#2563290238964912476.jpg b/benchmark/car_plate/image/#2563290238964912476.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..c923aca4029d083ca50a65691303fcd4d056331c
--- /dev/null
+++ b/benchmark/car_plate/image/#2563290238964912476.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d08610c72f9e957642f9c9b239003dac9686e1d2068e56213696aeadfb76bb6c
+size 210156
diff --git a/benchmark/car_plate/image/#286741634995118034.jpg b/benchmark/car_plate/image/#286741634995118034.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..6bc7614d7c8d4a41ea2c13a7e85b0fac652c6f71
Binary files /dev/null and b/benchmark/car_plate/image/#286741634995118034.jpg differ
diff --git a/benchmark/car_plate/image/#3334113938196704587.jpg b/benchmark/car_plate/image/#3334113938196704587.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..04cc996eef42c878b0d0852eaecb9c3c13ebe189
Binary files /dev/null and b/benchmark/car_plate/image/#3334113938196704587.jpg differ
diff --git a/benchmark/car_plate/image/#3517741965209098503.jpg b/benchmark/car_plate/image/#3517741965209098503.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b0ab674fbb250d7faf45271d8ee0c4bfb4171d08
Binary files /dev/null and b/benchmark/car_plate/image/#3517741965209098503.jpg differ
diff --git a/benchmark/car_plate/image/#3630149080300033894.jpg b/benchmark/car_plate/image/#3630149080300033894.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a51e2c1edc22e289ea9ab96675fae8c06d5bc804
Binary files /dev/null and b/benchmark/car_plate/image/#3630149080300033894.jpg differ
diff --git a/benchmark/car_plate/image/#6227411411065045336.jpg b/benchmark/car_plate/image/#6227411411065045336.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..62a0daea4721bfaf06dbfa4c5f6efd62314042ae
--- /dev/null
+++ b/benchmark/car_plate/image/#6227411411065045336.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4aaf83bf5d8d3f65246311256fe6662df4855c270f52e56898d1774203ece20f
+size 284278
diff --git a/benchmark/car_plate/image/#6238260831625012504.jpg b/benchmark/car_plate/image/#6238260831625012504.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e0e482eca0d7faf5e0f4c35e7ecd23c023e4012b
Binary files /dev/null and b/benchmark/car_plate/image/#6238260831625012504.jpg differ
diff --git a/benchmark/car_plate/image/#6659591284551536526.jpg b/benchmark/car_plate/image/#6659591284551536526.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..de07c302b45061b5a6e814fb153dfcc713abe115
Binary files /dev/null and b/benchmark/car_plate/image/#6659591284551536526.jpg differ
diff --git a/benchmark/car_plate/image/#7134505881093256854.jpg b/benchmark/car_plate/image/#7134505881093256854.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e7c2d0eae80bf775af75af16257d4df890fcd37e
Binary files /dev/null and b/benchmark/car_plate/image/#7134505881093256854.jpg differ
diff --git a/benchmark/car_plate/image/#7381771666275058687.jpg b/benchmark/car_plate/image/#7381771666275058687.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..1151c2ca7fd7e143003cfc6f55b223e91ca53d43
Binary files /dev/null and b/benchmark/car_plate/image/#7381771666275058687.jpg differ
diff --git a/benchmark/car_plate/image/#7518111569611972375.jpg b/benchmark/car_plate/image/#7518111569611972375.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..c8482a47e3cf0e0c442d3b4cc34436cfcb8f8dda
Binary files /dev/null and b/benchmark/car_plate/image/#7518111569611972375.jpg differ
diff --git a/benchmark/car_plate/image/#761631743535467855.jpg b/benchmark/car_plate/image/#761631743535467855.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ba277d38c0d230e1c0cbfc20611a9f67385038cd
--- /dev/null
+++ b/benchmark/car_plate/image/#761631743535467855.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03abfbef3345dfba5cb77d8634149c63d87c960e3296b857226da5c10253992b
+size 105318
diff --git a/benchmark/car_plate/image/#8529343536891984474.jpg b/benchmark/car_plate/image/#8529343536891984474.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f2b762c982476a1efdfea6980b6d4f401f0658a5
Binary files /dev/null and b/benchmark/car_plate/image/#8529343536891984474.jpg differ
diff --git a/benchmark/car_plate/image/#8813093836125699391.jpg b/benchmark/car_plate/image/#8813093836125699391.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..68973c6b245a4b3db7dd649cc70554d152c374b5
Binary files /dev/null and b/benchmark/car_plate/image/#8813093836125699391.jpg differ
diff --git a/benchmark/car_plate/image/#9170169492945964687.jpg b/benchmark/car_plate/image/#9170169492945964687.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..9e213a62d5ab989d50fbdeebe57b3ac88f52767d
Binary files /dev/null and b/benchmark/car_plate/image/#9170169492945964687.jpg differ
diff --git a/benchmark/car_plate/image/1140623623523992149.jpg b/benchmark/car_plate/image/1140623623523992149.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e6881cee359a2d53a15cce6eb38c20202dfff0af
Binary files /dev/null and b/benchmark/car_plate/image/1140623623523992149.jpg differ
diff --git a/benchmark/car_plate/image/1671412636281891010.jpg b/benchmark/car_plate/image/1671412636281891010.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8a42550b62d1f1444efa86b30e90b35b6a0aa588
Binary files /dev/null and b/benchmark/car_plate/image/1671412636281891010.jpg differ
diff --git a/benchmark/car_plate/image/1794121684277643246.jpg b/benchmark/car_plate/image/1794121684277643246.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..1489c1c974114d47b775c7ffd8bc244faf72e10f
--- /dev/null
+++ b/benchmark/car_plate/image/1794121684277643246.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62d439a4203741fb10190a9d8d2dbc0042a6caf2983eda01349037a4bbdfb0d6
+size 178045
diff --git a/benchmark/car_plate/image/1850854682574030615.jpg b/benchmark/car_plate/image/1850854682574030615.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..189ccedbc9f47a60d67c102fe683b07164863a26
Binary files /dev/null and b/benchmark/car_plate/image/1850854682574030615.jpg differ
diff --git a/benchmark/car_plate/image/2579392287500286667.jpg b/benchmark/car_plate/image/2579392287500286667.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e13b31944923f356d5634ecfe9f8c9cc0915644c
Binary files /dev/null and b/benchmark/car_plate/image/2579392287500286667.jpg differ
diff --git a/benchmark/car_plate/image/3021668729827686673.jpg b/benchmark/car_plate/image/3021668729827686673.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ec4d71fb45be1a60d01b7afcd0177eae96495582
--- /dev/null
+++ b/benchmark/car_plate/image/3021668729827686673.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3863e59fadee89c5680d94476e04b7bfa5aacd601ec39dc8c346b5edc13428b9
+size 544737
diff --git a/benchmark/car_plate/image/318568126668983171.jpg b/benchmark/car_plate/image/318568126668983171.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..6cca8e5e12ef4974985b6d0f90eb1098384273af
--- /dev/null
+++ b/benchmark/car_plate/image/318568126668983171.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb31a1a8b70ad4a0a2dba1623546d5c91aa4fbdbacce26e542ba2f1516f52b67
+size 1417832
diff --git a/benchmark/car_plate/image/3488762191409319384.jpg b/benchmark/car_plate/image/3488762191409319384.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..0c5f2248c27111686d82fc68d22e246eb14130c2
--- /dev/null
+++ b/benchmark/car_plate/image/3488762191409319384.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9305a22ee838cd83c6397f6be940354d284d0e86995d51816b8e8437ad4ca278
+size 100070
diff --git a/benchmark/car_plate/image/438344865304307623.jpg b/benchmark/car_plate/image/438344865304307623.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..0dcd67521fd48c129677551248c9de13cda764ef
Binary files /dev/null and b/benchmark/car_plate/image/438344865304307623.jpg differ
diff --git a/benchmark/car_plate/image/4966998296728060233.jpg b/benchmark/car_plate/image/4966998296728060233.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..cfd95fcd7dfa084253560f9e74fdb7628e76468c
Binary files /dev/null and b/benchmark/car_plate/image/4966998296728060233.jpg differ
diff --git a/benchmark/car_plate/image/5321180974886966504.jpg b/benchmark/car_plate/image/5321180974886966504.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f1240f514701a6e09a97118d4a39a0e6ce17e1df
Binary files /dev/null and b/benchmark/car_plate/image/5321180974886966504.jpg differ
diff --git a/benchmark/car_plate/image/582186526753952006.jpg b/benchmark/car_plate/image/582186526753952006.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..393fc53aed6f27a639f0f88a1dd3e9aa0b604380
Binary files /dev/null and b/benchmark/car_plate/image/582186526753952006.jpg differ
diff --git a/benchmark/car_plate/image/6386760609565435452.jpg b/benchmark/car_plate/image/6386760609565435452.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..733d116a0eaf18079ecbb5d39b98c1cbffaccb9a
--- /dev/null
+++ b/benchmark/car_plate/image/6386760609565435452.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ab62e1eada9e1562e3e37aba847a8fcdabe5210e11f5d73aafdd37b1ff91f6b
+size 151766
diff --git a/benchmark/car_plate/image_description.jsonl b/benchmark/car_plate/image_description.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..3ca5b2c9b4857a92de305e8e3c5ba7ee3e0bc438
--- /dev/null
+++ b/benchmark/car_plate/image_description.jsonl
@@ -0,0 +1,30 @@
+{"image_path": "image/#9170169492945964687.jpg", "image_description": "The image shows a white car with its headlights on, emitting a bright light that appears to be pinkish due to the camera's exposure settings or lighting conditions. The car has a visible license plate at the front, which reads \"ABE-8332.\" The car's grille and part of the bumper are also visible. The background is dark, suggesting the photo was taken at night."}
+{"image_path": "image/4966998296728060233.jpg", "image_description": "The image shows the front view of a white BMW car with its headlights on. The car's license plate is clearly visible and reads \"AJQ 3268.\" The lighting in the image appears to be predominantly purple, which might be due to a filter or special lighting effect applied to the photo. The car's grille and the BMW logo are also visible, indicating it is a BMW model. The overall scene suggests that the car is parked or moving at a low speed, as there are no visible signs of motion blur."}
+{"image_path": "image/#2563290238964912476.jpg", "image_description": "The image shows a blue public bus parked on the side of a road. The bus has multiple windows along its side and a digital display at the back showing \"03\" and \"Huangpu\". There are advertisements on the side and rear of the bus, including one for a bank and another for a shopping platform with various logos. The bus appears to be a CNG (Compressed Natural Gas) vehicle, as indicated by the label on the back. The license plate number visible on the bus is \"60212\". The bus is parked near a white railing, and there is a tree and some pavement in the background."}
+{"image_path": "image/1671412636281891010.jpg", "image_description": "The image shows a white Range Rover Evoque parked on a road with a clear view of its front. The vehicle has a sleek design with a black grille and silver accents around the headlights and bumper. The license plate is visible and reads \"B FK718.\" The car appears to be in good condition, and there are no other vehicles or significant objects in the immediate vicinity. The road markings indicate that the car is parked in a designated area."}
+{"image_path": "image/3021668729827686673.jpg", "image_description": "The image shows a white van driving on a road with clear lane markings. The van appears to be a minivan or small passenger van, and it has a visible license plate at the front. The license plate reads \"粤A-D6746,\" which suggests that the vehicle is registered in Guangdong Province, China (as indicated by the \"粤\" character). The van is moving forward, and there are two people visible inside, one of whom appears to be the driver. The lighting in the image is bright, indicating it might be taken during the day under sunny conditions."}
+{"image_path": "image/#3334113938196704587.jpg", "image_description": "The image shows a close-up view of the front section of a white Nissan vehicle parked on a street. The license plate is clearly visible and reads \"皖K·G199X.\" The car appears to be a modern model with a sleek design, featuring a prominent grille with the Nissan logo at the center. The headlights are also visible, and the overall condition of the car looks clean and well-maintained. The background includes part of another vehicle's rear end, suggesting that the photo was taken in a parking area or on a road where multiple vehicles are present."}
+{"image_path": "image/#6659591284551536526.jpg", "image_description": "The image shows the rear view of a bus with a blue and green color scheme. The bus has a digital display at the top showing some text, which appears to be in Chinese characters. Below the display, there is a license plate with the number \"81D002\" visible. The bus also features several red lights and a logo that reads \"BYD,\" indicating it is a BYD (Build Your Dreams) brand vehicle, known for producing electric vehicles. The bus's model number \"170371\" is displayed on the lower left side of the rear."}
+{"image_path": "image/1794121684277643246.jpg", "image_description": "The image shows the front view of a white and red bus parked at what appears to be a bus station or terminal. The bus has a license plate with the characters \"D 26226\" visible on the front. The bus model number \"ZK6115WD\" is also displayed on the front. There is a sign on the windshield that reads \"迎宾车,\" which translates to \"Welcome Car\" in English, suggesting it might be used for special events or tours. The bus is equipped with windshield wipers, and there are some reflections and shadows on the windshield. The surrounding area includes a metal railing and some yellow and black striped barriers, indicating a controlled access point."}
+{"image_path": "image/#1500244263428553553.jpg", "image_description": "The image shows a close-up view of a white Kia car parked on a wet surface, likely after rain. The car's front grille and part of its headlights are visible. The license plate of the car is blue with white characters, and it reads \"A82307.\" The car appears to be in good condition, and there are no visible damages or significant wear. The wet ground suggests recent rainfall, and the lighting indicates that the photo was taken during daylight hours."}
+{"image_path": "image/#8813093836125699391.jpg", "image_description": "The image shows the rear view of a silver Volkswagen car parked on a snowy surface. The car has a license plate with the characters \"京M R3801\" displayed in white on a blue background. The snow appears to be covering the roof and part of the back of the car, indicating recent snowfall or ongoing snow. The car's taillights are visible, and there is a small amount of snow on the ground around the vehicle. The background includes a building with a door and some windows partially covered by snow."}
+{"image_path": "image/#2249067055978906958.jpg", "image_description": "The image shows the front section of a white car, focusing on the lower part of the vehicle's grille and the license plate area. The license plate is blue with white characters and reads \"A-D0798.\" The car appears to be parked on a surface that looks like asphalt or concrete. The grille has a silver finish with horizontal slats. There is some greenery visible on the left side of the image, suggesting the car might be parked near a tree or bush."}
+{"image_path": "image/582186526753952006.jpg", "image_description": "The image shows the rear view of a beige-colored car parked on a gravel surface. The car has a distinctive rear spoiler and a badge with \"UAA\" on it, indicating it might be a model from the UAZ (Ulyanovsk Automobile Plant) brand. The license plate at the back reads \"京H 96171,\" which suggests that the vehicle is registered in Beijing, China. The car appears to be an older model, possibly from the late 20th century, given its design and condition. There are some items scattered around the car, including what looks like a piece of cloth or fabric on the ground near the rear wheel."}
+{"image_path": "image/#3517741965209098503.jpg", "image_description": "The image shows the front view of a Mazda vehicle with its headlights on. The license plate is clearly visible and reads \"BBS-0695.\" The car appears to be parked or moving at a slow speed, as indicated by the slight blur around the headlights. The environment suggests it might be nighttime or in a dimly lit area, given the bright headlights illuminating the surroundings. The car's grille and emblem are also visible, confirming it is a Mazda model."}
+{"image_path": "image/6386760609565435452.jpg", "image_description": "The image shows the front view of a large vehicle, likely a truck or bus, with a driver visible through the windshield. The vehicle has a dark blue exterior and features a prominent grille with red and white reflective stripes. The license plate is located at the bottom center of the image and reads \"新B 32122.\" The plate appears to be yellow with black characters, which is typical for certain regions in China. The vehicle also has additional lights on either side of the license plate area. The driver inside the vehicle is partially visible, wearing a light-colored shirt and holding what appears to be a phone or similar object. The windshield wipers are extended, suggesting that it might have rained recently or is currently raining."}
+{"image_path": "image/318568126668983171.jpg", "image_description": "The image shows the rear view of a bus with a green and red color scheme. The bus has a digital display at the back showing \"B-2099\" and some other information. There is an advertisement on the back of the bus that reads \"带租约销售 惠动全城\" which translates to \"Sales with rent agreement, benefiting the whole city.\" Below the advertisement, there is a phone number \"208 0000\" and a license plate number \"33133.\" The bus appears to be on a road with a metal railing beside it."}
+{"image_path": "image/5321180974886966504.jpg", "image_description": "The image shows a close-up view of a vehicle's front, specifically focusing on its headlights and part of the license plate. The time displayed in the top left corner of the image is 9:50. The license plate is partially visible and reads \"BBZ-2217.\" The vehicle appears to be a dark-colored sedan or hatchback. The headlights are bright, with one headlight being particularly overexposed due to the camera's exposure settings, making it difficult to see details clearly. The overall lighting suggests that the photo was taken at night or in low-light conditions."}
+{"image_path": "image/#7381771666275058687.jpg", "image_description": "The image shows a dark scene with a vehicle's headlights on, indicating it is nighttime. The vehicle appears to be stationary or moving very slowly, as the headlights are bright and focused directly towards the camera. The license plate of the vehicle is visible and reads \"京H·A9200,\" which suggests that the vehicle is registered in Beijing, China (as indicated by the \"京\" character). The background is mostly dark, with no other vehicles or significant objects discernible due to the low light conditions."}
+{"image_path": "image/#7518111569611972375.jpg", "image_description": "The image shows the front view of a car with its headlights on. The car appears to be a BMW, identifiable by the distinctive grille design. The license plate is visible and reads \"BEM-2398.\" The lighting from the headlights creates a bright glare, making it difficult to see details clearly. The car's bumper and part of the front fenders are also visible."}
+{"image_path": "image/1140623623523992149.jpg", "image_description": "The image shows the rear end of a white car with a visible license plate. The license plate is blue with white characters and reads \"A B2L00.\" The car appears to be parked on a paved surface, and part of the surrounding area, including some greenery, is visible in the background. The car's design suggests it might be a modern model, but specific details about the make and model are not discernible from this angle."}
+{"image_path": "image/2579392287500286667.jpg", "image_description": "The image shows a close-up view of a car's front, taken at night. The car's headlights are on, illuminating the area directly in front of it. The license plate is clearly visible and reads \"AGE-3802.\" The timestamp in the top left corner indicates that the photo was taken at 23:11:30. The car appears to be parked or moving very slowly, as there is no significant motion blur. The background is dark, suggesting it might be nighttime or in a dimly lit area."}
+{"image_path": "image/#6238260831625012504.jpg", "image_description": "The image shows the front view of a silver Toyota Corolla car. The car has a clear and visible license plate with the text \"京M 26227.\" The license plate is white with blue characters and numbers, which is typical for vehicles registered in Beijing, China. The car's headlights are on, and the grille features the Toyota logo prominently in the center. The vehicle appears to be parked outdoors, as indicated by the background elements like trees and a wall."}
+{"image_path": "image/#3630149080300033894.jpg", "image_description": "The image shows the rear view of a white Subaru vehicle parked on a street. The car has a sunroof and a rear window with a patterned design. The license plate is visible and reads \"辽B 7245U.\" The car appears to be a compact SUV model."}
+{"image_path": "image/1850854682574030615.jpg", "image_description": "The image shows a street scene with two cars parked on the side of the road. The focus is on the front of a white car, which has a visible license plate. The license plate reads \"皖L KX199,\" indicating that the vehicle is registered in Anhui Province, China. The car appears to be a Mazda model, as indicated by the logo on the grille. The background includes a building with a closed garage door and some greenery, including plants and flowers. The pavement is wet, suggesting recent rain."}
+{"image_path": "image/#7134505881093256854.jpg", "image_description": "The image shows the front grille of a car with a visible license plate. The license plate reads \"A-BON76\" and appears to be illuminated, possibly due to the lighting conditions or a reflective surface on the plate. The car's grille features the iconic four-ring logo, which is characteristic of Audi vehicles. The background is dark, suggesting that the photo was taken at night or in low-light conditions. There are some circular lights or reflections visible near the grille, which could be part of the car's design or external lighting."}
+{"image_path": "image/#761631743535467855.jpg", "image_description": "The image shows the rear end of a white Volkswagen vehicle. The license plate on the car reads \"A-P7Y78.\" The car appears to be parked, and part of another vehicle's taillight is visible in the foreground, indicating that the photo was taken from behind two vehicles. The lighting suggests it might be late afternoon or early evening, as the light has a warm tone."}
+{"image_path": "image/#8529343536891984474.jpg", "image_description": "The image shows a close-up view of a car's front grille and license plate area. The car appears to be a dark-colored sedan with its headlights on, emitting a bright light. The license plate is pink with black text that reads \"BLK-9801.\" The car's grille has a textured pattern, and there is a logo in the center of the grille, which looks like it could be from a luxury brand. The time displayed in the top left corner of the image is 5:16."}
+{"image_path": "image/3488762191409319384.jpg", "image_description": "The image shows the front view of a red truck. The truck has a logo on the front grille that appears to be a stylized \"F\" or similar design. Below the grille, there is a license plate with the characters \"苏N·E0852.\" The license plate is yellow with black text and numbers, which is typical for vehicles registered in China. The truck's headlights are visible, and the overall condition of the vehicle seems to be in good working order."}
+{"image_path": "image/438344865304307623.jpg", "image_description": "The image shows the rear view of a vehicle, specifically focusing on the license plate area. The license plate reads \"苏A·23158,\" which indicates that the vehicle is registered in Jiangsu Province, China. The vehicle appears to be a truck or a large van, as suggested by the size and design of the rear bumper and the presence of dual rear lights. The background includes some text and graphics on the vehicle's rear panel, but they are not clearly legible due to the angle and focus of the image. The overall condition of the image suggests it might have been taken from a moving vehicle or with a low-resolution camera."}
+{"image_path": "image/#6227411411065045336.jpg", "image_description": "The image shows the rear view of a bus with a digital display and various markings. The bus has a license plate at the bottom center that reads \"粤B 71481D.\" The bus also displays a phone number for bus inquiries: \"巴士热线: 89983563.\" The bus model is indicated as \"BYD,\" which stands for BYD (Build Your Dreams), a Chinese manufacturer known for electric vehicles. The bus has a yellow and white color scheme, and there is a slogan on the back that says \"预约出行 随需而至,\" which translates to \"Booked travel, according to your needs.\" The bus appears to be a BYD electric vehicle, as indicated by the logo and the design."}
+{"image_path": "image/#286741634995118034.jpg", "image_description": "The image shows the front view of a vehicle at night. The vehicle appears to be a bus or a large truck, as indicated by the size and design of the front grille and headlights. The brand name \"FOTON\" is visible on the front of the vehicle. The license plate is partially visible and reads \"豫PC1283,\" which suggests that the vehicle is registered in Henan Province, China (豫 is the abbreviation for Henan). The headlights are on, illuminating part of the front of the vehicle. The overall lighting is dim, indicating that the photo was taken in low-light conditions."}
diff --git a/benchmark/car_plate/user_input.txt b/benchmark/car_plate/user_input.txt
new file mode 100644
index 0000000000000000000000000000000000000000..838926286e7951cb40398db63e1aa20f57ffb009
--- /dev/null
+++ b/benchmark/car_plate/user_input.txt
@@ -0,0 +1 @@
+检测车牌号
\ No newline at end of file
diff --git "a/benchmark/car_plate/\345\217\202\350\200\203\345\233\276\347\211\207/4321d47e-31f9-4e68-b1da-9a6c855e67ea.jpg" "b/benchmark/car_plate/\345\217\202\350\200\203\345\233\276\347\211\207/4321d47e-31f9-4e68-b1da-9a6c855e67ea.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..e0e482eca0d7faf5e0f4c35e7ecd23c023e4012b
Binary files /dev/null and "b/benchmark/car_plate/\345\217\202\350\200\203\345\233\276\347\211\207/4321d47e-31f9-4e68-b1da-9a6c855e67ea.jpg" differ
diff --git "a/benchmark/car_plate/\345\217\202\350\200\203\345\233\276\347\211\207/8b798d6b-e581-4ed9-a1ba-c28f50a4702d.jpg" "b/benchmark/car_plate/\345\217\202\350\200\203\345\233\276\347\211\207/8b798d6b-e581-4ed9-a1ba-c28f50a4702d.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..a51e2c1edc22e289ea9ab96675fae8c06d5bc804
Binary files /dev/null and "b/benchmark/car_plate/\345\217\202\350\200\203\345\233\276\347\211\207/8b798d6b-e581-4ed9-a1ba-c28f50a4702d.jpg" differ
diff --git a/benchmark/fisherman/gt.jsonl b/benchmark/fisherman/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..18a0fa0c276e87023fb1c21e3313ec9ea0537f79
--- /dev/null
+++ b/benchmark/fisherman/gt.jsonl
@@ -0,0 +1,30 @@
+{"filename": "image/image##changjiang##2021_04_25_13_56_51_638-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [981.0768, 476.4646, 1038.3062, 560.7757], "label": 1}, {"is_ignored": false, "bbox": [1078.6734, 482.0853, 1133.8589, 576.616], "label": 1}, {"is_ignored": false, "bbox": [1494.7394, 473.812, 1563.7419, 570.6381], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_23_12_31_33_969-fishing.mp4##162.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [901.8014, 351.0855, 1104.1109, 681.8014000000001], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_26_10_41_23_352-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1255.4542, 643.322, 1443.7142999999999, 949.7918999999999], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_25_09_16_12_276-fishing.mp4##158.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1119.549, 318.542, 1351.6985, 652.6355], "label": 1}, {"is_ignored": false, "bbox": [1029.7173, 227.7008, 1214.4276, 612.2615], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_26_10_53_31_282-fishing.mp4##74.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1044.21, 376.2922, 1255.9267, 809.0598], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_01_08_50_50_652-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [893.7746, 539.1594, 937.6329999999999, 624.4397], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_23_16_34_50_593-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [649.133, 447.7693, 721.2051, 543.0074], "label": 1}, {"is_ignored": false, "bbox": [890.1987, 466.5684, 954.3262, 558.6853], "label": 1}]}
+{"filename": "image/image##changjiang##2021_06_19_13_01_47_221-fishing.mp4##34.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1012.245, 366.8622, 1176.1297, 696.5225], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_26_11_04_01_928-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [583.0385, 359.5307, 635.4341, 452.9493], "label": 1}, {"is_ignored": false, "bbox": [799.932, 454.5739, 855.9831, 537.0259], "label": 1}, {"is_ignored": false, "bbox": [897.0061, 484.2242, 950.6202, 560.5837], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_24_06_13_29_399-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [865.9916, 490.1389, 978.1117999999999, 693.4137], "label": 1}]}
+{"filename": "image/image##chowder##江堤达标纪念碑东侧高倍球机_1600249854808.mp4##102.jpg", "image_height": 1440, "image_width": 2560, "instances": [{"is_ignored": false, "bbox": [376.4169, 592.807, 411.9643, 705.6106], "label": 1}, {"is_ignored": false, "bbox": [522.6474, 591.0833, 563.0418, 703.6385], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_25_17_03_08_137-fishing.mp4##169.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [830.5349, 300.6378, 997.0545999999999, 577.2712], "label": 1}, {"is_ignored": false, "bbox": [226.9949, 159.6657, 434.1731, 489.0657], "label": 1}, {"is_ignored": false, "bbox": [0.0, 164.3397, 151.8865, 446.1809], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_23_08_41_12_169-fishing.mp4##245.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [695.4271, 378.5427, 977.6382, 666.1809000000001], "label": 1}, {"is_ignored": false, "bbox": [405.0754, 335.1256, 700.8543, 652.613], "label": 1}]}
+{"filename": "image/image##chowder##浒通河闸西侧普通球机_1600327196011.mp4##1.jpg", "image_height": 576, "image_width": 704, "instances": [{"is_ignored": false, "bbox": [320.0383, 261.3816, 358.8088, 366.2381], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_17_09_28_14_597-fishing.mp4##145.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [795.9815, 374.3414, 1032.2848, 829.8864000000001], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_08_15_23_31_150-fishing.mp4##75.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [894.381, 451.84, 985.016, 720.8586], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_26_09_38_00_142-fishing.mp4##70.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [869.454, 349.515, 969.3329, 663.4201], "label": 1}, {"is_ignored": false, "bbox": [415.2428, 431.5584, 584.0857, 637.8559], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_23_11_21_29_168-fishing.mp4##217.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [879.9497, 400.2513, 1092.9648, 681.1056000000001], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_17_08_55_21_721-fishing.mp4##80.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [849.0832, 498.5346, 1057.1515, 818.3174], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_23_07_33_18_206-fishing.mp4##163.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [500.0249, 348.3437, 707.1482, 676.5131], "label": 1}, {"is_ignored": false, "bbox": [721.9427, 369.863, 973.4495, 679.203], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_23_05_59_27_157-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [950.4653, 453.8498, 1002.8207, 607.8363], "label": 1}, {"is_ignored": false, "bbox": [229.9622, 449.7864, 312.2054, 549.6191], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_25_09_33_38_445-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [953.453, 563.6753, 992.9856, 637.7989], "label": 1}, {"is_ignored": false, "bbox": [902.8969, 509.318, 948.5114, 635.8983], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_08_11_53_34_002-fishing.mp4##69.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [908.0184, 519.2576, 1138.2207, 844.077], "label": 1}, {"is_ignored": false, "bbox": [1216.2554, 515.3558, 1444.5068, 851.8804], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_22_17_04_01_726-fishing.mp4##76.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [900.4124, 382.371, 1035.6803, 823.6656], "label": 1}, {"is_ignored": false, "bbox": [350.6254, 380.5233, 507.6027, 822.8591], "label": 1}]}
+{"filename": "image/image##changjiang##2021_04_23_18_34_53_717-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [139.8975, 423.3243, 274.7961, 601.7661], "label": 1}, {"is_ignored": false, "bbox": [638.937, 449.7917, 739.2572, 622.2570000000001], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_08_12_39_58_887-fishing.mp4##1.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1088.54, 528.4097, 1204.1447, 702.0907000000001], "label": 1}, {"is_ignored": false, "bbox": [935.6788, 501.0152, 1020.0538, 693.8723], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_08_10_25_40_742-fishing.mp4##137.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [905.3772, 525.8644, 1192.5855000000001, 902.7495000000001], "label": 1}, {"is_ignored": false, "bbox": [1270.1438, 500.4155, 1552.5047, 919.7153000000001], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_11_06_30_17_609-fishing.mp4##129.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [847.4043, 467.6376, 1024.8642, 786.0215000000001], "label": 1}]}
+{"filename": "image/image##changjiang##2021_06_24_15_42_29_196-fishing.mp4##62.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [799.2102, 362.4675, 1032.4307, 646.1141], "label": 1}]}
+{"filename": "image/image##changjiang##2021_05_07_08_38_51_400-fishing.mp4##88.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1030.8286, 370.8718, 1145.6802, 781.3380999999999], "label": 1}]}
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_22_17_04_01_726-fishing.mp4##76.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_22_17_04_01_726-fishing.mp4##76.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..17755600f48c85d65091375b4cdc81c637eeaf77
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_22_17_04_01_726-fishing.mp4##76.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cff7f167a20a5517d0a376ed7426b50aa1cc34247d1ab204c8189446c68d5c09
+size 1158972
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_23_05_59_27_157-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_23_05_59_27_157-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..41c02aabbbe77c963cddb2e70feca278aaf80c66
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_23_05_59_27_157-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f42c6396378df644a666da7943f8a018f5c71208771b32bd3183b8c1ee7a0a11
+size 825270
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_23_07_33_18_206-fishing.mp4##163.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_23_07_33_18_206-fishing.mp4##163.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..46b2f222181c02f3c62e45c2692ff3ef5168b7b9
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_23_07_33_18_206-fishing.mp4##163.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58cb36786d4febbdbd49679fc4b2c68d58095df10b6570c56f929da52735efdd
+size 1024932
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_23_08_41_12_169-fishing.mp4##245.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_23_08_41_12_169-fishing.mp4##245.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..cd6faf6728d7c6edda3439a4f5a9c565b2907d62
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_23_08_41_12_169-fishing.mp4##245.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7e61fc688620463907a2dfc3ca2bcdd8b145dd44050648d9d0dce0f279a2a78
+size 1085985
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_23_11_21_29_168-fishing.mp4##217.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_23_11_21_29_168-fishing.mp4##217.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..2cd5ee4e0d3b1089df4a0c81e0c682589c8f2d27
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_23_11_21_29_168-fishing.mp4##217.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97759389559e5669e1f56c8b57dd88c5b8f631beab52e8dd3050fb3009e85302
+size 1047632
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_23_12_31_33_969-fishing.mp4##162.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_23_12_31_33_969-fishing.mp4##162.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..08ae1a90941a49f19ccd3bb4a6e39d21d12bef77
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_23_12_31_33_969-fishing.mp4##162.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97e1bad9e29e4e8e7895edc898b351a489eee6a08bc28b2c7b98e52acde638e1
+size 951212
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_23_16_34_50_593-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_23_16_34_50_593-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..2ab7105bb107099d412db167309765e4819ac7a7
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_23_16_34_50_593-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5a9eea02d97f71cde6584df193bbb66482e2f0cb83ccd3c1df393f867860e16
+size 892750
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_23_18_34_53_717-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_23_18_34_53_717-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ef308e179f82058caac0d06c409427b5354d7afd
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_23_18_34_53_717-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b90c00c52c4f6d491289c18a4fac935a549eb5a42689fa3e2c7a3e41885e509
+size 516676
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_24_06_13_29_399-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_24_06_13_29_399-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ccc2f8da7967d588dbfe3febf004d3b732cc5148
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_24_06_13_29_399-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f00739f03b451c2eb6221790e53f8db2021293cda72d8308a496f964e573a6dc
+size 456841
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_25_09_16_12_276-fishing.mp4##158.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_25_09_16_12_276-fishing.mp4##158.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..9b73219fb4885dffc4dde421265348b230af8578
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_25_09_16_12_276-fishing.mp4##158.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e50ef1d0cec510cc24b617884cd5e2d45ca68f7567c823aaa18feeae27ec356
+size 925205
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_25_09_33_38_445-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_25_09_33_38_445-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a6d7df0b46fa18ad51a0bb0eec470ee627b7e5d3
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_25_09_33_38_445-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74926fb1300a0b704c8be175d8264ec1c6815a5b9e709c2197704982d27accff
+size 814256
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_25_13_56_51_638-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_25_13_56_51_638-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..5b8954f6c6db214c39953dfe054637d32522a88f
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_25_13_56_51_638-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fd7bf2b9674d7099d3bd38d59ff7c379356881652bd6634a23ee2d2160b5591
+size 735534
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_25_17_03_08_137-fishing.mp4##169.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_25_17_03_08_137-fishing.mp4##169.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..45249803c527603571eb3319df73ce0e7b5214fd
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_25_17_03_08_137-fishing.mp4##169.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3edbfe74f462fff12915673de8a39e9b08aa50dbc1798a3c1e321f1fd7f9ac3d
+size 909327
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_26_09_38_00_142-fishing.mp4##70.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_26_09_38_00_142-fishing.mp4##70.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3fc6bac6d6e84fd24b8a59f2f366c96eb23a5674
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_26_09_38_00_142-fishing.mp4##70.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4c39029cf467f5239301e4556b6b181ded88654750d7e591f1ec50a620db7ec
+size 612926
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_26_10_41_23_352-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_26_10_41_23_352-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a889e04a652e73708c2294115250a45437ce512f
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_26_10_41_23_352-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:968d8eb3a8385263cd4d1afea89ae2d9ea56284a41e83073cd2cb5ca6ec7ef4f
+size 679641
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_26_10_53_31_282-fishing.mp4##74.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_26_10_53_31_282-fishing.mp4##74.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e9b9a02f3ff08466e3348a412d37ffd06fc8a66b
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_26_10_53_31_282-fishing.mp4##74.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbc4ff55d9f53fff5c94cf0068408aa7b85c6c90d23114b389dc1994bb7dec47
+size 999817
diff --git a/benchmark/fisherman/image/image##changjiang##2021_04_26_11_04_01_928-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_04_26_11_04_01_928-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..03db1d6ea7392cca0c8b74ea6ee755c5498e7aae
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_04_26_11_04_01_928-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b89f02ad11e183fa3d76e06848608bac45f1e9a5e40a5f3c0fa8d3ca8393f8a0
+size 902431
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_01_08_50_50_652-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_01_08_50_50_652-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b483c84f3742af8ccc3928ebe6987dfe2710f352
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_01_08_50_50_652-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d1d0991d837cd27612983f0ef9460d30a1d5cac7fefce5949e56bed43f8854a
+size 1082853
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_07_08_38_51_400-fishing.mp4##88.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_07_08_38_51_400-fishing.mp4##88.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f2688a8289b13bac19d5078db432b687319523e8
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_07_08_38_51_400-fishing.mp4##88.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:317ef5fb408f92a238f5aebb8ffda060fb4a8843e3e1c885cb796ca0888683f7
+size 581990
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_08_10_25_40_742-fishing.mp4##137.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_08_10_25_40_742-fishing.mp4##137.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..eed2593a791e0912254bc29682891037b7cbff43
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_08_10_25_40_742-fishing.mp4##137.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaa1b2136e7ac2a82d365e5730afa95fed6bf12d464d2cd319b0c220ca44eb47
+size 941579
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_08_11_53_34_002-fishing.mp4##69.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_08_11_53_34_002-fishing.mp4##69.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..55ec6d3345f772455e58816dd1f523717c573307
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_08_11_53_34_002-fishing.mp4##69.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa1be3595182ef7a7103c49d010990bc4d15b1b483e64383c7da2aa2fe3ca8cf
+size 1058394
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_08_12_39_58_887-fishing.mp4##1.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_08_12_39_58_887-fishing.mp4##1.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..573ddd82be5a628634752f515ea9f67c7be735c3
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_08_12_39_58_887-fishing.mp4##1.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2f22d696d8b0541612067b880ab5432cc6cf2131bbad2e62dffc9f9f9433926
+size 664837
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_08_15_23_31_150-fishing.mp4##75.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_08_15_23_31_150-fishing.mp4##75.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b245eb5de1fec92698b2f559dcffa7cf1e0f708f
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_08_15_23_31_150-fishing.mp4##75.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d67336df956fa06325433219b836fff495599c0cb1e9deb99b4665843e67f9cd
+size 946984
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_11_06_30_17_609-fishing.mp4##129.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_11_06_30_17_609-fishing.mp4##129.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..c7a90363033bfc1ae8e8698d1e47804b871550a0
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_11_06_30_17_609-fishing.mp4##129.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f560ecdd346c84bb53cee80998d400414b34f11d0b3bc56703652b2141c2541f
+size 1098167
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_17_08_55_21_721-fishing.mp4##80.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_17_08_55_21_721-fishing.mp4##80.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3d8cf55ab2de63b2f53f2a417c9ab168c696fe6c
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_17_08_55_21_721-fishing.mp4##80.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07fcab070d0adb9f0d449bd88ae5b2e0c29a7d665c123f4031041179273eae20
+size 991764
diff --git a/benchmark/fisherman/image/image##changjiang##2021_05_17_09_28_14_597-fishing.mp4##145.jpg b/benchmark/fisherman/image/image##changjiang##2021_05_17_09_28_14_597-fishing.mp4##145.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f26b95d87bad46cf14e2724b1816c14feaa14d34
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_05_17_09_28_14_597-fishing.mp4##145.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5ae890ad597a82cafc259fbc88cef0f9631d60a2de2450a15cee7dfb9b3278a
+size 1099888
diff --git a/benchmark/fisherman/image/image##changjiang##2021_06_19_13_01_47_221-fishing.mp4##34.jpg b/benchmark/fisherman/image/image##changjiang##2021_06_19_13_01_47_221-fishing.mp4##34.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..7419dbd996849a775c9316531c532b99d7b0fee9
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_06_19_13_01_47_221-fishing.mp4##34.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:679fd432925f6296ec8847ba3076536a1245783b622520d35d2451cdec20ebc8
+size 804626
diff --git a/benchmark/fisherman/image/image##changjiang##2021_06_24_15_42_29_196-fishing.mp4##62.jpg b/benchmark/fisherman/image/image##changjiang##2021_06_24_15_42_29_196-fishing.mp4##62.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..abd1c8acfc4c7d26aaa7ed9ef08d3834ee3f0f3e
--- /dev/null
+++ b/benchmark/fisherman/image/image##changjiang##2021_06_24_15_42_29_196-fishing.mp4##62.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7c6ac423157d4b70c1eddfd71f6787c3aa987c8f9c0b9a3fc6242c5195295e9
+size 1119547
diff --git "a/benchmark/fisherman/image/image##chowder##\346\261\237\345\240\244\350\276\276\346\240\207\347\272\252\345\277\265\347\242\221\344\270\234\344\276\247\351\253\230\345\200\215\347\220\203\346\234\272_1600249854808.mp4##102.jpg" "b/benchmark/fisherman/image/image##chowder##\346\261\237\345\240\244\350\276\276\346\240\207\347\272\252\345\277\265\347\242\221\344\270\234\344\276\247\351\253\230\345\200\215\347\220\203\346\234\272_1600249854808.mp4##102.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..3d37ccc044bfd16fc433be32c0286adcd5b39d4f
--- /dev/null
+++ "b/benchmark/fisherman/image/image##chowder##\346\261\237\345\240\244\350\276\276\346\240\207\347\272\252\345\277\265\347\242\221\344\270\234\344\276\247\351\253\230\345\200\215\347\220\203\346\234\272_1600249854808.mp4##102.jpg"
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3ffdbbffb752b4147f1a331f5320f0dcc49025d4c6fe062b5dc01cc0abfeb89
+size 434376
diff --git "a/benchmark/fisherman/image/image##chowder##\346\265\222\351\200\232\346\262\263\351\227\270\350\245\277\344\276\247\346\231\256\351\200\232\347\220\203\346\234\272_1600327196011.mp4##1.jpg" "b/benchmark/fisherman/image/image##chowder##\346\265\222\351\200\232\346\262\263\351\227\270\350\245\277\344\276\247\346\231\256\351\200\232\347\220\203\346\234\272_1600327196011.mp4##1.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..c835457fe4caffb8a5f49d0b200c6e5510663ab0
--- /dev/null
+++ "b/benchmark/fisherman/image/image##chowder##\346\265\222\351\200\232\346\262\263\351\227\270\350\245\277\344\276\247\346\231\256\351\200\232\347\220\203\346\234\272_1600327196011.mp4##1.jpg"
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc4c32b575de3e6f2ad0d6fa4cf59805782e4ffd31346bd10ad986ebc2a05155
+size 152140
diff --git a/benchmark/fisherman/image_description.jsonl b/benchmark/fisherman/image_description.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..ece84291ec51c5bd15cd709d4adb32ff7155c831
--- /dev/null
+++ b/benchmark/fisherman/image_description.jsonl
@@ -0,0 +1,30 @@
+{"image_path": "image/image##changjiang##2021_04_25_13_56_51_638-fishing.mp4##1.jpg", "image_description": "The image shows a group of people fishing on a riverbank. The date and time stamp in the top right corner indicates that the photo was taken on April 25, 2021, at 13:56:50. The scene appears to be overcast, as suggested by the muted lighting and the presence of what looks like raindrops on the water's surface.\nThere are four individuals visible in the image, each sitting or standing near the edge of the river with fishing rods. One person in the center is holding an umbrella, likely for protection against the rain. The riverbank is lined with green vegetation, and the water is calm but has some ripples, indicating a gentle flow. The overall atmosphere suggests a peaceful, rainy day spent fishing."}
+{"image_path": "image/image##changjiang##2021_04_23_12_31_33_969-fishing.mp4##162.jpg", "image_description": "The image shows a person fishing by a body of water. The individual is seated on a small, portable chair or stool near the edge of the water. They are wearing a red long-sleeve shirt and dark pants. The person appears to be holding a fishing rod and is engaged in the act of fishing. There is a yellow bag with a logo on it placed next to them, along with a black object that could be a cooler or another piece of fishing equipment. The surrounding area includes green vegetation and some bare soil. The timestamp in the top right corner indicates the date and time as April 23, 2021, at 12:31:39 PM. The water is calm, reflecting the person and the vegetation."}
+{"image_path": "image/image##changjiang##2021_04_26_10_41_23_352-fishing.mp4##1.jpg", "image_description": "The image shows two individuals in a rural setting near a body of water, likely a pond or small river, surrounded by green vegetation. The person standing appears to be wearing a dark jacket and holding a yellow object, possibly a tool or equipment related to fishing or agriculture. The individual sitting on a small stool near the water seems to be engaged in some activity involving the water, potentially fishing or examining something in the water. The time stamp on the image indicates it was taken on April 26, 2021, at 10:41:23 AM. The text \"可见光\" (visible light) is visible in the bottom left corner, which might indicate that the image is from a security camera or similar surveillance system."}
+{"image_path": "image/image##changjiang##2021_04_25_09_16_12_276-fishing.mp4##158.jpg", "image_description": "The image shows two individuals fishing by a riverbank. The scene appears to be during the daytime, as indicated by the timestamp in the top right corner which reads \"2021-04-25 09:16:16 星期日\" (Sunday, April 25, 2021). The water level is relatively low, and there are some aquatic plants visible in the water. The riverbank is lined with green vegetation.\nThe two individuals are seated on chairs or stools that are partially submerged in the water. They are both holding fishing rods and appear to be engaged in fishing. One person is wearing a dark jacket and a white cap, while the other is dressed in a lighter-colored jacket and a darker cap. There are also some items placed on the ground near them, including what looks like a cooler box and possibly some fishing gear.\nThe overall setting suggests a calm and peaceful environment, likely a recreational fishing spot."}
+{"image_path": "image/image##changjiang##2021_04_26_10_53_31_282-fishing.mp4##74.jpg", "image_description": "The image shows a person standing on the edge of a body of water, likely a river or a pond, surrounded by green vegetation. The individual appears to be engaged in fishing, as they are holding a fishing rod and seem to be adjusting their clothing. The person is wearing dark clothing, which includes a jacket and pants, and is standing in shallow water. The timestamp in the top right corner indicates that the photo was taken on April 26, 2021, at 10:53:32 AM. The environment looks calm, with no other people visible in the immediate vicinity."}
+{"image_path": "image/image##changjiang##2021_05_01_08_50_50_652-fishing.mp4##1.jpg", "image_description": "The image shows a river with a bridge crossing over it. The bridge has railings on both sides and appears to be made of metal. There is a person standing on the bridge, wearing dark clothing. On the right side of the image, near the riverbank, there is another person who seems to be fishing. This individual is wearing red clothing and is holding a fishing rod. The surrounding area includes green grass and some trees in the background. The timestamp on the image indicates that it was taken on May 1, 2021, at 8:50:49 AM."}
+{"image_path": "image/image##changjiang##2021_04_23_16_34_50_593-fishing.mp4##1.jpg", "image_description": "The image shows a riverbank scene with two individuals near the water's edge. The area is lush with green vegetation, and the river appears calm with gentle ripples on the surface. The timestamp in the top right corner indicates that the photo was taken on April 23, 2021, at 16:34:49.\nOne person is seated on the bank, while the other is standing closer to the water, appearing to be engaged in some activity near the riverbed. There is a yellow object on the ground near the standing individual, which could be a container or a tool related to their activity. The overall setting suggests a peaceful, natural environment, possibly a fishing spot or a place for leisurely activities by the river."}
+{"image_path": "image/image##changjiang##2021_06_19_13_01_47_221-fishing.mp4##34.jpg", "image_description": "The image shows a person fishing in a shallow body of water surrounded by tall grasses and reeds. The individual is seated on a small chair or platform, holding a fishing rod. They are wearing a white shirt and dark pants, and have a blue umbrella providing shade above them. The time stamp in the top right corner indicates that the photo was taken on June 19, 2021, at 13:01:44. The scene appears to be outdoors, likely in a natural setting such as a riverbank or a pond."}
+{"image_path": "image/image##changjiang##2021_04_26_11_04_01_928-fishing.mp4##1.jpg", "image_description": "The image shows a serene riverside scene with two individuals engaged in fishing activities. The river is calm, and the surrounding area is lush with green vegetation. The timestamp on the image indicates it was taken on April 26, 2021, at 11:04:01 AM.\nIn the foreground, there is a person sitting on the riverbank, holding a fishing rod, indicating they are actively fishing. This individual appears to be wearing dark clothing and is positioned near the water's edge. Another person is also visible further down the bank, crouching close to the water, possibly preparing or adjusting their fishing equipment. This second individual is also dressed in dark clothing and is closer to the center of the image.\nThe environment suggests a peaceful and natural setting, ideal for fishing. The vegetation along the riverbank is dense and green, providing a backdrop that enhances the tranquil atmosphere of the scene."}
+{"image_path": "image/image##changjiang##2021_04_24_06_13_29_399-fishing.mp4##1.jpg", "image_description": "The image shows a person fishing near a riverbank. The individual is seated on a rock or a small platform in the shallow water, holding a fishing rod. They appear to be wearing a dark-colored jacket and a light-colored face mask. There is a bicycle parked on the riverbank behind them, with some items placed next to it, possibly fishing gear or personal belongings. The surrounding area is lush with green vegetation, and the river appears calm. The timestamp in the top right corner indicates that the photo was taken on April 24, 2021, at 6:13:27 AM."}
+{"image_path": "image/image##chowder##江堤达标纪念碑东侧高倍球机_1600249854808.mp4##102.jpg", "image_description": "The image shows a river scene with several individuals standing on the rocky shore. The weather appears to be overcast or foggy, as the visibility is somewhat reduced. There are trees and grass along the riverbank, and the water looks calm.\nThere are at least four people visible in the image. They appear to be engaged in fishing activities, as they are holding fishing rods. One person is closer to the foreground on the left side, another is further back near the center, and two more are towards the right side of the image. The individuals are dressed in casual clothing suitable for outdoor activities.\nThe timestamp in the top right corner indicates that the photo was taken on September 16, 2020, at 17:41:32. The text at the bottom right corner seems to be location-specific information, possibly indicating the area where the photo was taken."}
+{"image_path": "image/image##changjiang##2021_04_25_17_03_08_137-fishing.mp4##169.jpg", "image_description": "The image shows a group of people fishing near a body of water. The scene appears to be outdoors, with greenery and a scooter parked on the bank. The time stamp on the image indicates it was taken on April 25, 2021, at 17:03:13.\nIn the foreground, there is a person sitting in the water, wearing a blue jacket, yellow gloves, and a red hat. This individual seems to be engaged in fishing activities, as they are holding a fishing rod and appear to be focused on something in the water. The water level is high enough that the person's legs and lower body are submerged.\nTo the left of the person in the water, another individual is standing on the bank, also holding a fishing rod. This person is wearing a dark jacket and a hat. There is a bucket placed on the ground next to them.\nIn the background, there is a scooter parked on the bank, with various items attached to it, including what looks like a basket or container. The overall setting suggests a casual fishing activity in a natural environment."}
+{"image_path": "image/image##changjiang##2021_04_23_08_41_12_169-fishing.mp4##245.jpg", "image_description": "The image shows two individuals sitting on chairs by a body of water, likely fishing. They are positioned near the edge of the water, with their fishing rods extended into the water. The person on the left is wearing a blue outfit and a white cap, while the person on the right is dressed in darker clothing and a white hat. There are various fishing-related items scattered around them, including bags and containers. The surrounding area includes greenery and some dirt patches. The timestamp on the image indicates it was taken on April 23, 2021, at 8:41:21 AM."}
+{"image_path": "image/image##chowder##浒通河闸西侧普通球机_1600327196011.mp4##1.jpg", "image_description": "The image shows a person standing near a body of water, likely a river or lake, on a cloudy day. The individual is wearing an orange safety vest and a white hat, which suggests they might be engaged in some form of work or activity that requires visibility and safety gear. The person appears to be facing away from the camera, looking towards the water.\nThe surrounding area includes tall grass and reeds, indicating a natural or semi-natural environment. There is also a metal pole with a yellow top visible in the foreground, possibly part of a monitoring or safety system. The timestamp on the image indicates it was taken on September 17, 2020, at 15:10:26.\nThe text in the image provides additional context:\n- \"南通海门开发区\" (Nantong Haimen Development Zone)\n- \"浒通河闸西侧普通球机\" (Ordinary ball machine west side of Hutan River sluice gate)\nThis suggests that the location is within the Nantong Haimen Development Zone, specifically near the west side of a sluice gate on the Hutan River. The mention of an \"ordinary ball machine\" could imply that the image is from a surveillance or monitoring system."}
+{"image_path": "image/image##changjiang##2021_05_17_09_28_14_597-fishing.mp4##145.jpg", "image_description": "The image shows a person standing in a grassy area near a body of water. The individual appears to be engaged in some activity related to fishing or handling something in the water. They are wearing a long-sleeved shirt and pants, and there is a white umbrella set up nearby, possibly for shade. The time stamp on the image indicates it was taken on May 17, 2021, at 9:28:21 AM. The scene suggests a rural or outdoor setting, likely a pond or small lake."}
+{"image_path": "image/image##changjiang##2021_05_08_15_23_31_150-fishing.mp4##75.jpg", "image_description": "The image shows a person standing in shallow water near a grassy area. The individual appears to be fishing, as they are holding a fishing rod and are standing in the water with their feet submerged up to the ankles. The person is wearing a light-colored shirt and dark shorts. The time stamp on the image indicates it was taken on May 8, 2021, at 15:23:27. The surrounding environment includes green grass and some small plants growing out of the water. The water level is relatively low, allowing the person to stand comfortably without being fully submerged."}
+{"image_path": "image/image##changjiang##2021_04_26_09_38_00_142-fishing.mp4##70.jpg", "image_description": "The image shows two individuals fishing by a riverbank on April 26, 2021, at 9:37:58 AM. The person on the left is sitting and appears to be holding a fishing rod, while the person on the right is standing and also holding a fishing rod. Both individuals are wearing dark clothing suitable for fishing. The background consists of green vegetation and a sandy riverbank. The water in the river is calm with some ripples visible. The timestamp indicates that the photo was taken during daylight hours."}
+{"image_path": "image/image##changjiang##2021_04_23_11_21_29_168-fishing.mp4##217.jpg", "image_description": "The image shows a person fishing by a body of water. The individual is wearing a red long-sleeve shirt and appears to be bending over, possibly adjusting or handling something on the fishing rod. There is a yellow bag and a black object, which could be a cooler or another piece of fishing equipment, placed on the ground near the person. The surrounding area includes green vegetation and some bare soil. The timestamp in the top right corner indicates that the photo was taken on April 23, 2021, at 11:21:36 AM. The water is calm, reflecting the person and the surroundings."}
+{"image_path": "image/image##changjiang##2021_05_17_08_55_21_721-fishing.mp4##80.jpg", "image_description": "The image shows a person fishing in a grassy area near a body of water. The individual is seated and appears to be engaged in fishing activities. They are using a fishing rod and have an umbrella set up for shade. The person is wearing a dark jacket and a hat. The surrounding environment is lush with tall green grass, and there is another umbrella visible on the left side of the image. The timestamp in the top right corner indicates that the photo was taken on May 17, 2021, at 8:55:25 AM."}
+{"image_path": "image/image##changjiang##2021_04_23_07_33_18_206-fishing.mp4##163.jpg", "image_description": "The image shows two individuals sitting on folding chairs by a body of water, likely fishing. They are both wearing hats and appear to be engaged in their activity. The person on the left is holding a fishing rod and seems to be looking at something in their hands, possibly a phone or a small object. The person on the right is also holding a fishing rod and appears to be looking down at it.\nThe setting is outdoors, with greenery visible in the background and some items placed on the ground near the individuals, including what looks like a bag and some other equipment. The time stamp on the image indicates that the photo was taken on April 23, 2021, at 7:33 AM. The weather appears to be overcast, as suggested by the lighting in the image."}
+{"image_path": "image/image##changjiang##2021_04_23_05_59_27_157-fishing.mp4##1.jpg", "image_description": "The image shows a riverbank scene with two individuals engaged in fishing activities. The time and date stamp on the image indicates it was taken on April 23, 2021, at 5:59:26 AM.\nIn the foreground, there is a person wearing bright orange clothing, standing in the shallow water near the riverbank. This individual appears to be holding a fishing rod, suggesting they are actively fishing. The person's reflection can be seen in the water, indicating that the water is calm.\nIn the background, another person is seated on the riverbank, also engaged in fishing. This individual is wearing darker clothing and is positioned closer to the edge of the bank where a bicycle is parked. The surrounding area is lush with green vegetation, and the river appears relatively calm with some ripples on the surface.\nThe overall setting suggests a peaceful early morning activity by the river."}
+{"image_path": "image/image##changjiang##2021_04_25_09_33_38_445-fishing.mp4##1.jpg", "image_description": "The image shows a river with a concrete structure on the left side that appears to be a dam or a spillway. There is a staircase leading up to a platform at the top of the structure. The riverbank on the right side is grassy and there is a dirt path running parallel to the river.\nIn the foreground, near the riverbank, there are two individuals who appear to be fishing. They are standing close to the water's edge, each holding a fishing rod. The person on the left is wearing dark clothing, while the person on the right is also dressed in dark clothing. Both individuals seem to be focused on their fishing activity.\nThe timestamp in the image indicates it was taken on April 25, 2021, at 9:33:35 AM. The date and time suggest it might be early morning. The overall scene gives a peaceful impression of a quiet fishing spot."}
+{"image_path": "image/image##changjiang##2021_05_08_11_53_34_002-fishing.mp4##69.jpg", "image_description": "The image shows two individuals sitting on the edge of a body of water, likely fishing. They are using umbrellas for shade, and there is a motorbike parked nearby. The scene appears to be outdoors, with grassy areas surrounding the water. The timestamp on the image indicates it was taken on May 8, 2021, at 11:53:34 AM.\nThe target fisherman is the individual on the left side of the image. This person is wearing a blue shirt and appears to be engaged in some activity related to fishing, possibly handling a fishing rod or net. The fisherman has a red bucket next to them, which might be used for collecting fish or other items. The umbrella they are using is blue and white."}
+{"image_path": "image/image##changjiang##2021_04_22_17_04_01_726-fishing.mp4##76.jpg", "image_description": "The image shows two individuals standing near a body of water, likely a river or stream, surrounded by tall green vegetation. The time and date stamp on the image indicates it was taken on April 22, 2021, at 17:04:01. Both individuals appear to be engaged in fishing activities, as evidenced by the fishing rods visible in the foreground.\nThe person on the left is wearing a dark jacket, light-colored shirt, and dark pants. They have a fishing rod in their hand and seem to be adjusting something on it. The person on the right is dressed in a dark blazer, light-colored shirt, and dark pants. They are standing with their hands behind their back, observing the surroundings.\nThere is also a white bucket on the ground near the person on the right, which could be used for holding fish or other fishing equipment. The overall setting suggests a peaceful outdoor activity, possibly a leisurely fishing trip."}
+{"image_path": "image/image##changjiang##2021_04_23_18_34_53_717-fishing.mp4##1.jpg", "image_description": "The image shows a group of people fishing near a body of water. The scene appears to be during the evening, as indicated by the timestamp \"2021-04-23 18:34:51 星期五\" (Friday, April 23, 2021, 6:34 PM). There are three individuals visible in the image:\n1. **Person on the left**: This individual is sitting on the edge of the water, holding a fishing rod and appears to be actively fishing.\n2. **Person in the middle**: This person is standing and seems to be observing or assisting with the fishing activity.\n3. **Person on the right**: This individual is also sitting on the edge of the water, holding a fishing rod and appears to be engaged in fishing.\nThe background consists of greenery and a sandy bank, suggesting a natural setting. The water is calm, and there are some objects like a red bag and a yellow object near the individuals, possibly related to their fishing equipment. The overall atmosphere suggests a peaceful and leisurely activity."}
+{"image_path": "image/image##changjiang##2021_05_08_12_39_58_887-fishing.mp4##1.jpg", "image_description": "The image shows two people sitting on the bank of a river under umbrellas, likely fishing. There are two motorbikes parked nearby, one black and one red. The area is surrounded by green vegetation, indicating a natural setting. The timestamp on the image indicates it was taken on May 8, 2021, at 12:39:56 PM.\nThe target fisherman appears to be the person on the right side of the image, wearing a white shirt and dark pants. They are seated on a chair with a fishing rod in front of them, and there is a red bucket next to them. The umbrella they are using is white. The other person, who seems to be wearing a darker outfit, is also seated but facing away from the camera, making it difficult to determine their specific actions or attire."}
+{"image_path": "image/image##changjiang##2021_05_08_10_25_40_742-fishing.mp4##137.jpg", "image_description": "The image shows two individuals sitting on a dirt path near a body of water, likely fishing. The person on the left is wearing a blue jacket and appears to be focused on something in their hands, possibly related to fishing equipment. The person on the right is wearing a white shirt and is seated under a white umbrella for shade. There is a red bag and some other fishing-related items next to them. In the background, there is tall green grass and a parked electric scooter with a basket containing what looks like a child's stroller or carrier. The timestamp on the image indicates it was taken on May 8, 2021, at 10:25:43 AM."}
+{"image_path": "image/image##changjiang##2021_05_11_06_30_17_609-fishing.mp4##129.jpg", "image_description": "The image shows a person fishing by a riverbank. The individual is seated on a small chair and appears to be holding a fishing rod. They are wearing a blue jacket, brown pants, and a white cap. There is a white bucket next to them, possibly containing fishing equipment or bait. The riverbank is lined with grass, and there are some items scattered around, including what looks like a black bag and a yellow object. The timestamp in the top right corner indicates that the photo was taken on May 11, 2021, at 6:30 AM. The water in the river is calm, and the overall scene suggests a peaceful morning fishing activity."}
+{"image_path": "image/image##changjiang##2021_06_24_15_42_29_196-fishing.mp4##62.jpg", "image_description": "The image shows a person fishing in a shallow body of water. The individual is wearing a wide-brimmed hat and appears to be crouched down on a small platform or float in the water. They are holding a fishing rod with both hands, and there are some items placed on the ground near the water's edge, including what looks like a red bag and other miscellaneous objects. The surrounding area is grassy, and the time stamp in the top right corner indicates that the photo was taken on June 24, 2021, at 15:42:30."}
+{"image_path": "image/image##changjiang##2021_05_07_08_38_51_400-fishing.mp4##88.jpg", "image_description": "The image shows a person standing on a muddy bank next to a body of water, likely a river or canal. The individual appears to be engaged in some form of activity involving the water, possibly fishing or cleaning. They are holding a long pole or stick, which they are extending into the water. The person is wearing a light-colored jacket and dark pants. The surrounding area is lush with green vegetation, including tall grasses and reeds. The time stamp in the top right corner indicates that the photo was taken on May 7, 2021, at 8:38:52 AM. The text \"可见光\" (visible light) is visible in the bottom left corner, suggesting that the image might have been captured using a thermal imaging camera."}
diff --git a/benchmark/fisherman/user_input.txt b/benchmark/fisherman/user_input.txt
new file mode 100644
index 0000000000000000000000000000000000000000..1eb3d384dcf5d81119681a891bb621be36294792
--- /dev/null
+++ b/benchmark/fisherman/user_input.txt
@@ -0,0 +1 @@
+检测河边钓鱼的人
\ No newline at end of file
diff --git "a/benchmark/fisherman/\345\217\202\350\200\203\345\233\276\347\211\207/\345\236\202\351\222\223\344\272\272\345\221\230\346\243\200\346\265\21320251026_130031_3.jpg" "b/benchmark/fisherman/\345\217\202\350\200\203\345\233\276\347\211\207/\345\236\202\351\222\223\344\272\272\345\221\230\346\243\200\346\265\21320251026_130031_3.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..c963d55f15b0acdf52d51c440939798de525a53d
--- /dev/null
+++ "b/benchmark/fisherman/\345\217\202\350\200\203\345\233\276\347\211\207/\345\236\202\351\222\223\344\272\272\345\221\230\346\243\200\346\265\21320251026_130031_3.jpg"
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9f721248f10291ac07281ad44a2dc404e4c5f19eba26e6d41d3bf93ad70bd3a
+size 185119
diff --git a/benchmark/garbage_dump/gt.jsonl b/benchmark/garbage_dump/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..28b93f5a1b1fcce44c06bf9047af32b989f8dfb3
--- /dev/null
+++ b/benchmark/garbage_dump/gt.jsonl
@@ -0,0 +1,30 @@
+{"filename": "image/image##images##354.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [135.75101488497975, 182.13216712507565, 1071.9822788506804, 973.3152909336941], "label": 1}]}
+{"filename": "image/image##images##153.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [197.13125845737488, 187.06359945872802, 1123.6806495263872, 957.2395128552098], "label": 1}]}
+{"filename": "image/image##images##562.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [0, 319.4616700594264, 546.905766186699, 1036.7836679340278], "label": 1}, {"is_ignored": false, "bbox": [833.3570209090826, 224.55851153894997, 1117.1692143218365, 603.9208100673311], "label": 1}, {"is_ignored": false, "bbox": [665.0453459237183, 146.87748961145104, 792.3907067484604, 316.6713040444407], "label": 1}]}
+{"filename": "image/image##images##911.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [994.1999999999999, 471.61249694824215, 1513.8, 894.0124969482422], "label": 1}]}
+{"filename": "image/image##images##766.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [624.0749999999999, 374.4767826625279, 791.4857142857143, 533.1821398053851], "label": 1}]}
+{"filename": "image/image##images##1114.jpg", "image_height": 576, "image_width": 704, "instances": [{"is_ignored": false, "bbox": [317.5111111111111, 128.00740559895834, 543.6444444444444, 408.8962944878473], "label": 1}]}
+{"filename": "image/image##images##205.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [319.89174560216514, 175.3721244925575, 1174.830852503383, 860.7848443843031], "label": 1}]}
+{"filename": "image/image##images##902.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [343.8, 284.41249694824216, 803.4, 882.0124969482422], "label": 1}, {"is_ignored": false, "bbox": [865.8, 217.21249694824218, 1115.3999999999999, 534.0124969482422], "label": 1}]}
+{"filename": "image/image##images##68.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [653.8516683138686, 367.1321983880622, 842.3281226074974, 563.0878770584224], "label": 1}]}
+{"filename": "image/image##images##533.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [497.3366834170854, 225.22613065326632, 821.608040201005, 645.8291457286432], "label": 1}, {"is_ignored": false, "bbox": [834.7962805718344, 185.0942598239232, 1003.0788290482886, 404.9834564998235], "label": 1}]}
+{"filename": "image/image##images##681.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [420, 668.894472361809, 1920.0000000000002, 1080.0], "label": 1}]}
+{"filename": "image/image##images##27.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [642.3714680865066, 237.70437458057174, 912.8108099776656, 661.4252207205103], "label": 1}]}
+{"filename": "image/image##images##246.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [299.43166441136674, 185.6021650879567, 1076.9147496617052, 824.2489851150203], "label": 1}]}
+{"filename": "image/image##images##663.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [772.4939974103843, 290.9825344725278, 976.6803579775685, 566.2695027372138], "label": 1}]}
+{"filename": "image/image##images##103.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [603.107525107525, 367.6130450972065, 749.7741917741916, 523.168600652762], "label": 1}]}
+{"filename": "image/image##images##279.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [856.87018842499, 323.6655690683267, 1346.0041605819235, 933.8626401223856], "label": 1}]}
+{"filename": "image/image##images##934.jpg", "image_height": 1200, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1557.4594594594594, 531.5093809935424, 1919.9999999999995, 1200.0], "label": 1}]}
+{"filename": "image/image##images##647.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [535.3266331658291, 579.3467336683417, 1797.13567839196, 1080.0], "label": 1}]}
+{"filename": "image/image##images##626.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [533.9698492462311, 259.1457286432161, 884.0201005025125, 791.0050251256282], "label": 1}]}
+{"filename": "image/image##images##403.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [650.6532663316583, 495.22613065326635, 1087.537688442211, 963.3165829145729], "label": 1}]}
+{"filename": "image/image##images##901.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [682.1999999999999, 475.2124969482422, 1087.8, 889.2124969482422], "label": 1}]}
+{"filename": "image/image##images##1026.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [795.3333333333334, 676.0138854980469, 1155.3333333333335, 1080.0], "label": 1}]}
+{"filename": "image/image##images##1142.jpg", "image_height": 576, "image_width": 704, "instances": [{"is_ignored": false, "bbox": [312.53333333333336, 113.074072265625, 541.5111111111112, 412.4518500434028], "label": 1}]}
+{"filename": "image/image##images##651.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [543.4673366834171, 320.2010050251256, 852.8140703517588, 656.6834170854271], "label": 1}]}
+{"filename": "image/image##images##596.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [479.69849246231155, 417.8894472361809, 964.070351758794, 1080.0], "label": 1}]}
+{"filename": "image/image##images##960.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [873.1933130699089, 283.5042902613122, 1015.4425531914894, 447.6380288631359], "label": 1}]}
+{"filename": "image/image##images##1412.jpg", "image_height": 2160, "image_width": 3840, "instances": [{"is_ignored": false, "bbox": [892.610247160133, 234.46858068845074, 1361.2559744257396, 435.8001386112204], "label": 1}]}
+{"filename": "image/image##images##45.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [856.078230450022, 219.30476683623695, 1103.1072648862137, 537.4344089699304], "label": 1}]}
+{"filename": "image/image##images##842.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [502.2, 304.8124969482422, 892.2, 847.2124969482422], "label": 1}, {"is_ignored": false, "bbox": [862.9232142857145, 197.4714255196708, 1048.4142857142858, 433.18571123395645], "label": 1}]}
+{"filename": "image/image##images##53.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [946.4321608040201, 506.08040201005025, 1620.7537688442212, 1039.2964824120602], "label": 1}]}
diff --git a/benchmark/garbage_dump/image/image##images##1026.jpg b/benchmark/garbage_dump/image/image##images##1026.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..d13b301ca69c67d20c443cc878427b904ebfdf46
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##1026.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60eb1ba47c4a8d4a88c293d60c899ba0eb5e95a3eccbef08cea58ff56c549d05
+size 283583
diff --git a/benchmark/garbage_dump/image/image##images##103.jpg b/benchmark/garbage_dump/image/image##images##103.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..afa9b16a5f6d1cd76f2691c865a8f1406a8a7aa9
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##103.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d53afe0dc265620f887a46997b3daf3cf24ec407103434c5f6c532acbe0ccf9a
+size 281369
diff --git a/benchmark/garbage_dump/image/image##images##1114.jpg b/benchmark/garbage_dump/image/image##images##1114.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8f8c66c8fd1b9eb8ef20ebb67b678040ba916378
Binary files /dev/null and b/benchmark/garbage_dump/image/image##images##1114.jpg differ
diff --git a/benchmark/garbage_dump/image/image##images##1142.jpg b/benchmark/garbage_dump/image/image##images##1142.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..314c6ed4c87e1fd7587159d005c002b40b5adc69
Binary files /dev/null and b/benchmark/garbage_dump/image/image##images##1142.jpg differ
diff --git a/benchmark/garbage_dump/image/image##images##1412.jpg b/benchmark/garbage_dump/image/image##images##1412.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..2755e2e7e5e96e8f07ab615e02662a1e2df3945a
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##1412.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:483c0138e4583f81606e57f2a1ea8ed1643c79bd4eca8a4c6b7468ae50d3dbea
+size 739655
diff --git a/benchmark/garbage_dump/image/image##images##153.jpg b/benchmark/garbage_dump/image/image##images##153.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..7b17c1b88bbcc95b042070e8e99c0e01caad140d
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##153.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d8d4ea10b249c523c269d11052b2b24168898e28930c6c4feb3e3dc01dbbb6e
+size 265875
diff --git a/benchmark/garbage_dump/image/image##images##205.jpg b/benchmark/garbage_dump/image/image##images##205.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..16bb3d63aecc7e7ace96d32c350eaa96b9eb8d88
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##205.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd89cd7a13dd7021084e4fd97d7cc84dcf267aefef1055a32ede8f1cde1bc28a
+size 253117
diff --git a/benchmark/garbage_dump/image/image##images##246.jpg b/benchmark/garbage_dump/image/image##images##246.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..bf096ff76ea56097bd24569d0aadacbce7eda77a
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##246.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2203ca4ea0fd44e0bff2adf4b7a70b6c6de22ba3e1b7cbcdd7746ff811a5660e
+size 230343
diff --git a/benchmark/garbage_dump/image/image##images##27.jpg b/benchmark/garbage_dump/image/image##images##27.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a56fd9b2f77dd39291a6795fcf066878866ea2cf
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##27.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99a501417e544d11d6a41cafa8ed7cfd524d843dd35e7c2cf999e1c8d23169ea
+size 276605
diff --git a/benchmark/garbage_dump/image/image##images##279.jpg b/benchmark/garbage_dump/image/image##images##279.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e906d31c185e361a5b5fe654567f00df7efcaccc
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##279.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e769833f5ea7e9aa1584ef5c45a1bd268f6662962c12e0861074712e56caf493
+size 244221
diff --git a/benchmark/garbage_dump/image/image##images##354.jpg b/benchmark/garbage_dump/image/image##images##354.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..99a57ea9c6a7c5b9a4ee285e19be07770f16b76a
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##354.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e6e9968024af1aa84335607cf5e8b356b00a13d4c224c79c753b1bc74fabc05
+size 275970
diff --git a/benchmark/garbage_dump/image/image##images##403.jpg b/benchmark/garbage_dump/image/image##images##403.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..cdf7a6d6ba09aab48818d1b2b78a6a8fac91bd9a
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##403.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b3bbc9fb8d3fa9ee781ecde055ee889e214e17cc4669361f4e0b262c4e97dce
+size 289124
diff --git a/benchmark/garbage_dump/image/image##images##45.jpg b/benchmark/garbage_dump/image/image##images##45.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..87239570de9c274a27dafb3a3966fee9f68cd0c4
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##45.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2bdd5a065217041528e92656601323faf3d08eb2ec52f7a7b9894fced210dd4
+size 228631
diff --git a/benchmark/garbage_dump/image/image##images##53.jpg b/benchmark/garbage_dump/image/image##images##53.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..dcbebf6ddb533a8d59825c37a91d5785b6bc361f
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##53.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cee8a40d2f4ebc413f2f8d4dd496fa64c3eadfdcbe6f191d110fc44638c4d2ae
+size 257585
diff --git a/benchmark/garbage_dump/image/image##images##533.jpg b/benchmark/garbage_dump/image/image##images##533.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..da01f35c8bf11e9f47f30a889727e0de9cf6de91
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##533.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82211bce9e87a681f7840281908b2c3cfb136e599fcbc6d69553f72860790447
+size 251316
diff --git a/benchmark/garbage_dump/image/image##images##562.jpg b/benchmark/garbage_dump/image/image##images##562.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..0bf75a67db11d50f779124fb1d7461dc89b003b9
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##562.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:912bfc76f45b5dcc068e0d4f970d66213f122a656a71116b502cc04999a8fd97
+size 229556
diff --git a/benchmark/garbage_dump/image/image##images##596.jpg b/benchmark/garbage_dump/image/image##images##596.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4ea7d02ba35f4d64d9d7a549d004ec9feb0ecf00
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##596.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ee8959611f3b4821e952e0f6a6a1c404594ba2cc05901b21e8181c84be2d6ce
+size 256452
diff --git a/benchmark/garbage_dump/image/image##images##626.jpg b/benchmark/garbage_dump/image/image##images##626.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..9f5ce2d3cb7a1d40f2b2c275fb0b310a0182f4e1
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##626.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f6d4e986e32650c97002d3bce76981f7e5a3f8a53e6fcd786ee80b0228590d7
+size 251408
diff --git a/benchmark/garbage_dump/image/image##images##647.jpg b/benchmark/garbage_dump/image/image##images##647.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f47a74bd6555ba3c8e112d4adce40116178071db
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##647.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e538551cff60270a24f38bf853f944bfe61b901d32d942e31e84111f13c1887
+size 275195
diff --git a/benchmark/garbage_dump/image/image##images##651.jpg b/benchmark/garbage_dump/image/image##images##651.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ca686750ba5fd4450a61ccc0b82db35871702f7d
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##651.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d90c76d7105dc0d8a029901c95acf5b555da567e14a7c310ba8fb4386bcbabe7
+size 222550
diff --git a/benchmark/garbage_dump/image/image##images##663.jpg b/benchmark/garbage_dump/image/image##images##663.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a7019e602dc8f163cba4d23184c81884453b27a9
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##663.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1656b2e5fa615fbdb2e9f3da31bb1904ee6c38ececc6495c1c5a6a23b7f323cb
+size 313001
diff --git a/benchmark/garbage_dump/image/image##images##68.jpg b/benchmark/garbage_dump/image/image##images##68.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a622b9698fa22ee916160f12315322f7be22eb4c
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##68.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d411274ce1ce7d4827d24c4a87de7ae021fdf351a0946b97b4bd76736b8d12f
+size 283754
diff --git a/benchmark/garbage_dump/image/image##images##681.jpg b/benchmark/garbage_dump/image/image##images##681.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..fd90a5bed48f68ad02a06c35a51ac6cbd8a56990
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##681.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3537fcad22e2bf62bfa2592310f7b8fa519341545d345602a50b486e606a5585
+size 276413
diff --git a/benchmark/garbage_dump/image/image##images##766.jpg b/benchmark/garbage_dump/image/image##images##766.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..15da1575fb039cc7b90b01ad026194950b395f15
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##766.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1153b3fae63085a4ca189cff076b68c73081a59d40b5eaa74b84571f80772560
+size 286023
diff --git a/benchmark/garbage_dump/image/image##images##842.jpg b/benchmark/garbage_dump/image/image##images##842.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3d1bfd43bed5eb38e7547714886128e58b841a33
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##842.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df7e8d42f23d6eb22f816cac468c43c474e0fce613db85c783fe36b7c0343658
+size 251345
diff --git a/benchmark/garbage_dump/image/image##images##901.jpg b/benchmark/garbage_dump/image/image##images##901.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..454dff3ce90ada02db817252ac6fdbd17bd06b1a
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##901.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6d0101f22053202ab6697e93921654e74595c3756703c3a18d5527da355d1e7
+size 255942
diff --git a/benchmark/garbage_dump/image/image##images##902.jpg b/benchmark/garbage_dump/image/image##images##902.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..5cba6cccdcef786934ba0867a560c295ddfbc96a
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##902.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd0cee675509432b73eb46a93f4f4fa27b15ffccbeb38381292f02de5913e2d1
+size 256320
diff --git a/benchmark/garbage_dump/image/image##images##911.jpg b/benchmark/garbage_dump/image/image##images##911.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..40c653dbdfe56bc72d5491dcab603bf79b51a441
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##911.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7596c7b4ee27d5e0fff949c52cb4f92835492af571208816ac349e1bd67b0f81
+size 265636
diff --git a/benchmark/garbage_dump/image/image##images##934.jpg b/benchmark/garbage_dump/image/image##images##934.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8e470dc53af6d846604f45cfb3df52fc20dd6745
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##934.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8487f44d5801bf51fa3c05958097be4bbad57451941cdff4ff65fc4522405ca6
+size 393737
diff --git a/benchmark/garbage_dump/image/image##images##960.jpg b/benchmark/garbage_dump/image/image##images##960.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..6c4889b6a48f530cdb69466eea2b35541ebdcdc2
--- /dev/null
+++ b/benchmark/garbage_dump/image/image##images##960.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe7649358f0671a2e304c22d2ec2918c15d78cf3bbbfc73d3402e6781bb66027
+size 448758
diff --git a/benchmark/garbage_dump/image_description.jsonl b/benchmark/garbage_dump/image_description.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..40d7749db943492e53c34e56324241a608900ffe
--- /dev/null
+++ b/benchmark/garbage_dump/image_description.jsonl
@@ -0,0 +1,30 @@
+{"image_path": "image/image##images##354.jpg", "image_description": "The image appears to be a surveillance camera capture taken at night on August 16, 2021, at 00:34:46. The scene shows a wet road with visible reflections from streetlights and vehicle headlights, indicating recent rain. There is a large truck with a green tarp covering its cargo area driving across a crosswalk. The truck has a yellow license plate and some text on its side that is partially obscured. The surrounding environment includes parked cars, a pedestrian crossing, and a traffic light. The timestamp and other details suggest this is a security camera feed."}
+{"image_path": "image/image##images##153.jpg", "image_description": "The image shows a nighttime scene captured by a traffic camera at an intersection. The timestamp on the image indicates it was taken on August 13, 2021, at 00:06:42. The camera's identification number is \"32010564001310000618,\" and the location appears to be \"螺塘路天康街东北角\" (Luotang Road Tiankang Street Northeast Corner).\nIn the foreground, there is a large truck with a green tarp covering its cargo area. The truck has a license plate that reads \"苏A-19449.\" The truck is positioned near the pedestrian crossing, which is marked with white stripes. The road surface looks wet, possibly from recent rain, and there are some puddles visible.\nIn the background, there are several cars parked along the side of the road, and the surrounding area includes trees, streetlights, and a fence. The traffic lights are green, indicating that vehicles have the right of way. The overall scene suggests a quiet urban environment at night."}
+{"image_path": "image/image##images##562.jpg", "image_description": "The image shows a nighttime traffic scene captured by a surveillance camera. The timestamp on the image indicates it was taken on August 16, 2021, at 19:56:51. The location appears to be a roundabout or intersection with multiple lanes of traffic. There are several vehicles visible, including trucks and cars, with their headlights and taillights illuminated due to the low light conditions.\nIn the foreground, there is a large truck with a green tarp covering its cargo area. The license plate of the truck is partially visible, showing \"苏A·TX262.\" The truck is positioned in the right lane, moving away from the camera's perspective. The road surface looks wet, suggesting recent rain.\nThere is no specific target garbage dump identified in the image. The focus seems to be on the traffic flow and the general environment of the intersection."}
+{"image_path": "image/image##images##911.jpg", "image_description": "The image appears to be a nighttime surveillance camera capture of an urban street intersection. The timestamp on the image indicates it was taken on August 19, 2021, at 21:10:37. The scene shows a multi-lane road with vehicles, including a white car and a truck, visible. There are traffic lights and a pedestrian crossing in the center of the intersection. The surroundings include trees, buildings, and construction cranes, suggesting ongoing development or maintenance work.\nThere is no visible garbage dump in the image. The focus seems to be on the traffic and the urban environment rather than any specific object like a garbage dump. If you have any further questions or need more details about the image, feel free to ask!"}
+{"image_path": "image/image##images##766.jpg", "image_description": "The image shows a nighttime street scene captured by a surveillance camera. The timestamp on the image indicates it was taken on August 17, 2021, at 00:24:16. The location appears to be an intersection with a pedestrian crossing and traffic lights. There are several vehicles visible, including a truck with a flatbed trailer and a few cars parked or moving along the road.\nThe environment looks wet, suggesting recent rain, as there are puddles on the ground. The area around the intersection has some greenery, including trees and bushes. On the right side of the image, there is a construction site with a green fence and some construction equipment visible.\nThere is no visible garbage dump in the image. The focus seems to be on the traffic and the surrounding urban environment rather than any specific waste management facility."}
+{"image_path": "image/image##images##1114.jpg", "image_description": "The image shows a nighttime street scene with a green traffic light on the left side. There are two vehicles visible: one appears to be a bus or large vehicle on the right side of the road, and another vehicle is further ahead on the road. The road surface looks wet, possibly from recent rain, as there are reflections of the lights and the vehicle's headlights.\nThere is no visible garbage dump in the image. The focus is on the vehicles and the street environment. The timestamp at the top of the image indicates it was taken on August 29, 2021, at 20:34:55. The bottom right corner has some text that seems to indicate the location or source of the image, but it is not fully legible."}
+{"image_path": "image/image##images##205.jpg", "image_description": "The image appears to be a nighttime traffic surveillance camera capture from a street intersection. The timestamp on the image indicates it was taken on August 13, 2021, at 01:03:14. The license plate number of the vehicle in focus is partially visible and reads \"苏A 99845\". This vehicle is a large, dirty, and heavily used truck with a green tarp covering part of its cargo area. The truck is crossing a pedestrian crosswalk, and there are other vehicles visible in the background, including a white car and a blue van. The road surface looks wet, possibly due to recent rain, and there are some puddles visible. The surrounding environment includes streetlights, traffic signals, and some greenery along the roadside. There is no visible garbage dump in the image."}
+{"image_path": "image/image##images##902.jpg", "image_description": "The image shows a nighttime scene at an intersection with two trucks driving on a road. The timestamp indicates it was taken on August 20, 2021, at 00:48:51. The license plates on the trucks are visible and appear to be from China, as indicated by the characters on them.\nThe truck on the left has a license plate that reads \"苏A Z9676,\" which suggests it is registered in Jiangsu Province, China. The truck on the right has a license plate that reads \"苏A X9612.\"\nThe road appears to be well-lit, likely due to streetlights, and there are traffic lights visible in the background. The surrounding area includes greenery and what looks like a pedestrian crossing. The image also contains some text in Chinese, which seems to be related to the camera system or location details."}
+{"image_path": "image/image##images##68.jpg", "image_description": "The image shows a nighttime scene at an intersection with wet roads, indicating recent or ongoing rain. There are several vehicles visible, including a white truck with the license plate \"苏A 1X567\" and a red vehicle on the right side of the image. The traffic lights are visible, showing red for the direction the camera is facing. There are also some construction barriers and a green fence on the right side of the image, suggesting a construction site nearby.\nThere is no visible garbage dump in the image. The focus appears to be on the traffic and the wet conditions due to the rain."}
+{"image_path": "image/image##images##533.jpg", "image_description": "The image shows a nighttime traffic scene captured by a surveillance camera. The timestamp on the image indicates it was taken on August 12, 2021, at 19:58:10. The location appears to be an intersection with multiple lanes and traffic signals. There are several vehicles visible, including trucks and cars, moving through the intersection. The road markings are clearly visible, and there are pedestrian crossings on both sides of the road.\nThere is no visible garbage dump in the image. The focus of the image is on the traffic flow and the surrounding urban environment. The lighting suggests that it is nighttime, and the area is well-lit by streetlights. The image also includes some text and numbers, which appear to be related to the camera's identification and location details."}
+{"image_path": "image/image##images##681.jpg", "image_description": "The image shows a nighttime street scene captured by a surveillance camera. The timestamp on the image indicates it was taken on August 15, 2021, at 20:23:06. The location appears to be an intersection with multiple lanes and traffic lights. There are several vehicles visible, including a large truck in the foreground, which seems to be a dump truck given its design and the text on the side that includes \"8430\" and \"元前路友谊街西南角1,\" which translates to \"Yuanqian Road Friendship Street Southwest Corner 1.\" \nIn the background, there are buildings, some of which appear to be under construction as indicated by the presence of a crane and a green construction fence. There are also parked cars along the right side of the road.\nNo specific target garbage dump is clearly identifiable in the image. However, the large truck in the foreground could potentially be used for transporting garbage, but without additional context or clearer labeling, it cannot be confirmed as a garbage dump."}
+{"image_path": "image/image##images##27.jpg", "image_description": "The image shows a nighttime scene at an intersection captured by a surveillance camera. The timestamp on the image indicates it was taken on August 13, 2021, at 00:47:44. The license plate of the vehicle in the foreground reads \"苏A·X9076,\" which suggests that the vehicle is registered in Jiangsu Province, China.\nThe road appears to be wet, possibly from recent rain, and there are visible reflections of light on the surface. The traffic lights are green, indicating that vehicles have the right of way. There are no visible garbage dumps in the image. The surroundings include well-maintained landscaping with trees and shrubs, and the area is illuminated by streetlights. The image also contains some text in Chinese characters, which likely provides additional information about the location or the camera system used to capture the footage."}
+{"image_path": "image/image##images##246.jpg", "image_description": "The image appears to be a nighttime traffic surveillance camera capture from a street intersection. The timestamp on the image indicates it was taken on August 17, 2021, at 23:54:12. The scene shows a large truck with a green and white cargo area driving through an intersection. The truck's license plate is visible, and it seems to be a heavy-duty vehicle, possibly used for construction or transportation purposes.\nThere are no visible garbage dumps in the image. The focus is on the truck and its movement through the intersection. The surrounding environment includes other vehicles parked along the side of the road, pedestrian crosswalks, and traffic lights. The lighting suggests it is nighttime, and the area appears to be well-lit by streetlights."}
+{"image_path": "image/image##images##663.jpg", "image_description": "The image shows a nighttime scene at an intersection with wet roads reflecting the lights from vehicles and street lamps. The timestamp on the image indicates it was taken on August 13, 2021, at 22:10:34. There is a vehicle with the license plate \"苏AX1517\" driving away from the camera, its taillights glowing red. The road appears to be wet, suggesting recent rain. Traffic lights are visible, with one showing a red light and another a green light. The surroundings include trees and some greenery along the sides of the road. There are no visible garbage dumps in the image."}
+{"image_path": "image/image##images##103.jpg", "image_description": "The image shows a nighttime street scene captured by a surveillance camera. The timestamp on the image indicates it was taken on August 17, 2021, at 00:57:50. The location appears to be an intersection with a pedestrian crossing and traffic lights. There are several vehicles visible, including a large truck with a flatbed trailer and a smaller car. The road surface looks wet, suggesting recent rain.\nThere is no visible garbage dump in the image. The surroundings include trees lining the street, parked cars, and some industrial structures in the background. The image also contains some text in Chinese characters, which likely provides additional information about the camera's location or the event being recorded."}
+{"image_path": "image/image##images##279.jpg", "image_description": "The image appears to be a nighttime traffic surveillance camera capture from August 17, 2021, at 19:56:38. The scene shows an urban road intersection with multiple lanes and traffic signals. There are several vehicles visible, including a large truck in the foreground on the right side of the image, which seems to be a garbage collection vehicle based on its design and markings. The truck has a green top and a brown body with text that likely indicates it belongs to a waste management service. The license plate number \"8260\" is visible on the back of the truck. The surrounding area includes pedestrian crossings, trees, and other vehicles in the background. The timestamp and location information suggest this is a specific intersection in China."}
+{"image_path": "image/image##images##934.jpg", "image_description": "The image shows a street intersection with clear traffic markings and pedestrian crossings. The date and time stamp on the image indicates it was taken on August 18, 2021, at 08:49:45. There are trees lining both sides of the street, providing shade. A green truck is visible on the right side of the image, partially obstructing the view of the intersection. The road appears to be relatively empty, with only a few vehicles and bicycles visible.\nThere is no garbage dump present in the image. The focus is on the urban infrastructure and traffic setup at the intersection."}
+{"image_path": "image/image##images##647.jpg", "image_description": "The image shows a nighttime street scene captured by a surveillance camera. The timestamp on the image indicates it was taken on August 15, 2021, at 20:19:15. The scene includes a road intersection with traffic lights and several vehicles, including a large truck in the foreground with a green and white trailer. The truck has a license plate that reads \"苏AXQ723.\" There are also other vehicles, such as cars and motorcycles, visible on the road.\nIn the background, there are buildings and some construction activity, indicated by a crane and a green fence surrounding an area. The environment appears to be urban, with streetlights illuminating the scene.\nThere is no visible garbage dump in the image. The focus is on the road and the vehicles, with no specific mention or indication of a garbage dump being present."}
+{"image_path": "image/image##images##626.jpg", "image_description": "The image shows a nighttime traffic scene captured by a surveillance camera. The timestamp on the image indicates it was taken on August 13, 2021, at 20:13:10. The location appears to be a roundabout or intersection with multiple lanes of traffic. The road surface is wet, suggesting recent rain, and there are reflections from vehicle headlights and taillights.\nIn the foreground, there is a large truck with the license plate \"苏AX1844\" (indicating it is registered in Jiangsu Province, China). The truck is moving away from the camera, and its taillights are illuminated. The truck's bed is open, and it seems to be carrying some cargo, though the details of the cargo are not clear.\nThere are other vehicles visible in the background, including cars and another truck, all with their headlights on due to the low light conditions. The surrounding area includes trees and some greenery, as well as streetlights that illuminate the scene. The overall atmosphere is one of a busy urban environment during a rainy night."}
+{"image_path": "image/image##images##403.jpg", "image_description": "The image shows a nighttime scene captured by a surveillance camera on August 16, 2021, at 22:42:18. The timestamp and camera ID are visible in the top left corner. The scene appears to be an urban street with wet pavement, indicating recent rain. There are several vehicles on the road, including a large white truck in the foreground, a smaller car behind it, and other vehicles further down the street. On the right side of the image, there is a construction site with green fencing and some construction equipment.\nThere is no visible garbage dump in the image. The focus seems to be on the traffic and the surrounding urban environment rather than any specific waste management facility."}
+{"image_path": "image/image##images##901.jpg", "image_description": "The image appears to be a nighttime security camera capture from a street scene. The timestamp on the image indicates it was taken on August 20, 2021, at 00:54:55. The scene shows a road with a white truck driving towards the camera, its headlights illuminating the road ahead. There are several cars parked on the right side of the road, and construction equipment is visible in the background on the right, suggesting ongoing construction or maintenance work.\nThere is no visible garbage dump in the image. The focus seems to be on the traffic and the surrounding urban environment rather than any specific waste management facility."}
+{"image_path": "image/image##images##1026.jpg", "image_description": "The image appears to be a nighttime security camera capture of an urban street scene. The timestamp on the image indicates it was taken on August 20, 2021, at 4:19:42 AM. The scene shows a relatively quiet street with minimal traffic. There are a few vehicles visible, including a truck in the foreground and some cars parked or moving in the background.\nThe environment includes streetlights illuminating the road, trees lining the sidewalk, and buildings in the distance. A construction crane is visible on the right side of the image, suggesting ongoing construction work nearby.\nThere is no visible garbage dump in the image. The focus seems to be on the street and its immediate surroundings rather than any specific area for waste disposal."}
+{"image_path": "image/image##images##1142.jpg", "image_description": "The image shows a nighttime street scene with a green traffic light on the left side. There are two vehicles visible: a car and a bus, both approaching the camera. The bus has its headlights on, illuminating part of the road ahead. The timestamp at the top right corner indicates the photo was taken on August 30, 2021, at 21:58:59.\nThere is no visible garbage dump in the image. The scene appears to be an urban street with some buildings and streetlights in the background, but no specific structures resembling a garbage dump are present."}
+{"image_path": "image/image##images##651.jpg", "image_description": "The image shows a nighttime traffic scene at an intersection. The timestamp on the image indicates it was taken on August 17, 2021, at 22:53:47. The road markings and traffic lights are clearly visible, with vehicles moving along the road. There are two trucks visible in the image, one closer to the camera and another further ahead. The truck closer to the camera has a license plate that reads \"苏A·X1643.\" The surrounding area includes greenery and streetlights illuminating the scene.\nThere is no garbage dump present in the image. The focus is on the traffic and the intersection itself."}
+{"image_path": "image/image##images##596.jpg", "image_description": "The image shows a nighttime traffic scene at an intersection. The timestamp on the image indicates it was taken on August 12, 2021, at 20:38:05. The scene includes multiple vehicles, with a truck prominently visible in the foreground, facing away from the camera. The truck has a license plate that reads \"苏AX5298.\" There are also other vehicles on the road, including a red car and a white car further down the street.\nThe intersection is well-lit with streetlights, and there are traffic lights visible. The road markings indicate lanes for different directions of traffic. There are also pedestrian crossings marked on the road. The surrounding area appears to be landscaped with greenery and some trees.\nThere is no visible garbage dump in the image. The focus of the image is on the traffic and the intersection itself."}
+{"image_path": "image/image##images##960.jpg", "image_description": "The image shows a traffic intersection with multiple lanes and a pedestrian crossing. The date and time stamp on the image indicates it was taken on August 18, 2021, at 12:10:04. There are several vehicles visible, including a truck carrying what appears to be a load of sand or gravel. The road markings include directional arrows and lane dividers.\nThere is no visible garbage dump in the image. The focus is on the traffic and the intersection itself. If you have any specific questions about the vehicles or the road layout, feel free to ask!"}
+{"image_path": "image/image##images##1412.jpg", "image_description": "The image shows a busy urban intersection with multiple lanes of traffic. There are several cars and motorcycles visible, with some vehicles stopped at a red light while others are in motion. The road markings include pedestrian crosswalks and lane dividers. The timestamp on the image indicates it was taken on August 29, 2021, at 14:58:22.\nThere is no visible garbage dump in the image. The scene appears to be a typical city street with traffic flow and infrastructure for pedestrians and vehicles."}
+{"image_path": "image/image##images##45.jpg", "image_description": "The image shows a nighttime traffic scene at an intersection with multiple lanes and a roundabout. The timestamp on the image indicates it was taken on August 12, 2021, at 20:13:07. There are several vehicles visible, including a truck with the license plate \"BAX0076\" in the center lane, heading straight. Other vehicles are seen in the left and right lanes, some with their headlights on.\nThe road markings are clearly visible, with pedestrian crosswalks and lane dividers. The surrounding area includes greenery and streetlights illuminating the scene. There are no visible garbage dumps in the image. The bottom right corner of the image contains text that appears to be related to the camera or system used to capture the footage."}
+{"image_path": "image/image##images##842.jpg", "image_description": "The image appears to be a nighttime traffic surveillance camera capture from a street intersection. The timestamp on the image indicates it was taken on August 15, 2021, at 20:01:36. The license plate number \"苏AX1807\" is visible on one of the trucks.\nIn the foreground, there is a large truck with a green and white body, and the text \"8260-高庙路龙王大街路口东南\" is displayed, which likely refers to the location or route of the truck. The truck seems to be a garbage collection vehicle, as indicated by the text \"垃圾车\" (garbage truck) written on its side. The truck's bed is open, and it appears to be collecting or transporting waste.\nThe background shows other vehicles, including a white truck and a car, moving along the road. The traffic lights are green, indicating that the vehicles have the right of way. The surrounding area includes well-maintained sidewalks, trees, and roundabouts, suggesting an urban environment."}
+{"image_path": "image/image##images##53.jpg", "image_description": "The image appears to be a nighttime traffic surveillance camera capture taken on August 17, 2021, at 01:26:44. The scene shows an urban street intersection with several vehicles and some construction activity.\n### Description of the Image:\n- **Time and Date**: The timestamp indicates it was taken at 01:26:44 on August 17, 2021.\n- **Location**: The location is indicated as \"元通路友谊街西南角\" (Yuantong Road Friendship Street Southwest Corner).\n- **Weather**: It seems to be raining or has recently rained, as the road surface looks wet.\n- **Vehicles**:\n  - A large truck with a green tarp covering its cargo area is prominently visible in the foreground, heading towards the intersection.\n  - Other vehicles include a white sedan and a smaller truck further down the road.\n  - There is also a forklift on the left side of the image near the sidewalk.\n- **Traffic Lights**: Traffic lights are visible at the intersection, showing red signals.\n- **Construction Activity**: On the right side of the image, there is a construction site with a green fence and some construction equipment, including a crane.\n- **Surroundings**: Trees line the left side of the street, and buildings can be seen in the background.\n### Target Garbage Dump:\nThere is no visible garbage dump in the image. The image primarily captures a street scene with vehicles, a construction site, and a rainy night setting. If you have any specific details or areas you're interested in, feel free to ask!"}
diff --git a/benchmark/garbage_dump/user_input.txt b/benchmark/garbage_dump/user_input.txt
new file mode 100644
index 0000000000000000000000000000000000000000..c9eaa665072990ffbf534598d9a03d2bd8c07d13
--- /dev/null
+++ b/benchmark/garbage_dump/user_input.txt
@@ -0,0 +1 @@
+检测道路上的垃圾车
\ No newline at end of file
diff --git "a/benchmark/garbage_dump/\345\217\202\350\200\203\345\233\276\347\211\207/image##images##153.jpg" "b/benchmark/garbage_dump/\345\217\202\350\200\203\345\233\276\347\211\207/image##images##153.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..7b17c1b88bbcc95b042070e8e99c0e01caad140d
--- /dev/null
+++ "b/benchmark/garbage_dump/\345\217\202\350\200\203\345\233\276\347\211\207/image##images##153.jpg"
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d8d4ea10b249c523c269d11052b2b24168898e28930c6c4feb3e3dc01dbbb6e
+size 265875
diff --git a/benchmark/ped_bev/gt.jsonl b/benchmark/ped_bev/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..da8d382fb3a62586840c9f269c0371dab3a1c9d0
--- /dev/null
+++ b/benchmark/ped_bev/gt.jsonl
@@ -0,0 +1,30 @@
+{"filename": "image/image##dazuzhuang##8_0141.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [901, 136, 1029, 283], "is_ignored": false}, {"label": 1, "bbox": [1033, 408, 1126, 524], "is_ignored": false}, {"label": 1, "bbox": [1308, 372, 1437, 489], "is_ignored": false}, {"label": 1, "bbox": [916, 637, 1043, 746], "is_ignored": false}]}
+{"filename": "image/image##hl_2405##test##18_00232.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##dazuzhuang##11_0087.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1724, 471, 1904, 566], "is_ignored": false}, {"label": 1, "bbox": [938, 420, 1109, 582], "is_ignored": false}, {"label": 1, "bbox": [890, 612, 1024, 768], "is_ignored": false}, {"label": 1, "bbox": [583, 842, 747, 1032], "is_ignored": false}]}
+{"filename": "image/image##hl_2405##test##9_00076.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##hl_2405##test##5_00048.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##hl_2405##test##12_00046.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##zc_zhugangchejian_frames##300340.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1745, 802, 1818, 853], "is_ignored": false}]}
+{"filename": "image/image##hl_2405##test##11_00055.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [1022.5169, 534.2617, 1153.7365, 648.6102000000001], "label": 1}]}
+{"filename": "image/image##dazuzhuang##3_0180.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1415, 0, 1626, 181], "is_ignored": false}, {"label": 1, "bbox": [864, 0, 1083, 213], "is_ignored": false}, {"label": 1, "bbox": [756, 375, 931, 563], "is_ignored": false}]}
+{"filename": "image/image##dazuzhuang##9_0356.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [594, 0, 716, 166], "is_ignored": false}, {"label": 1, "bbox": [1233, 0, 1363, 80], "is_ignored": false}]}
+{"filename": "image/image##dazuzhuang_2##test##20_01290.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [648.8833, 0.03, 732.4801, 103.1921], "label": 1}, {"is_ignored": false, "bbox": [940.5829, 39.1604, 1086.4327, 155.3659], "label": 1}]}
+{"filename": "image/image##zc_zhugangchejian_frames##500243.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1117, 450, 1174, 511], "is_ignored": false}, {"label": 1, "bbox": [1345, 453, 1423, 512], "is_ignored": false}]}
+{"filename": "image/image##zc_zhugangchejian_frames##200286.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##dazuzhuang##9_0287.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##hl_2405##test##3_00333.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##dazuzhuang_2##test##20_01200.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [923.4129, 28.5136, 1052.4031, 136.0055], "label": 1}, {"is_ignored": false, "bbox": [800.7224, 500.4205, 897.2388, 631.9779], "label": 1}]}
+{"filename": "image/image##dazuzhuang_2##test##20_00480.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [721.2377, 877.1651, 830.7243, 957.8571000000001], "label": 1}, {"is_ignored": false, "bbox": [636.8627, 935.7588, 791.2154, 1043.9061], "label": 1}]}
+{"filename": "image/image##zc_zhugangchejian_frames##100231.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1165, 691, 1231, 793], "is_ignored": false}, {"label": 1, "bbox": [1353, 906, 1414, 957], "is_ignored": false}]}
+{"filename": "image/image##zc_zhugangchejian_frames##400520.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [781, 551, 847, 630], "is_ignored": false}, {"label": 1, "bbox": [821, 492, 887, 563], "is_ignored": false}, {"label": 1, "bbox": [1099, 442, 1152, 514], "is_ignored": false}, {"label": 1, "bbox": [1266, 836, 1352, 916], "is_ignored": false}]}
+{"filename": "image/image##zc_zhugangchejian_frames##500805.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1052, 759, 1154, 835], "is_ignored": false}]}
+{"filename": "image/image##hl_2405##test##13_00207.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"is_ignored": false, "bbox": [482.1282, 536.6349, 595.227, 702.2216000000001], "label": 1}]}
+{"filename": "image/image##zc_zhugangchejian_frames##500144.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1001, 826, 1081, 885], "is_ignored": false}, {"label": 1, "bbox": [1146, 744, 1219, 801], "is_ignored": false}]}
+{"filename": "image/image##zc_zhugangchejian_frames##500055.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##zc_zhugangchejian_frames##500283.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1459, 619, 1549, 678], "is_ignored": false}]}
+{"filename": "image/image##dazuzhuang##0_0272.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [390, 872, 547, 1035], "is_ignored": false}, {"label": 1, "bbox": [437, 0, 571, 104], "is_ignored": false}, {"label": 1, "bbox": [675, 0, 780, 135], "is_ignored": false}]}
+{"filename": "image/image##zc_zhugangchejian_frames##500202.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1040, 511, 1094, 564], "is_ignored": false}, {"label": 1, "bbox": [1131, 568, 1198, 625], "is_ignored": false}]}
+{"filename": "image/image##zc_zhugangchejian_frames##300471.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [1425, 904, 1529, 970], "is_ignored": false}, {"label": 1, "bbox": [1465, 744, 1560, 851], "is_ignored": false}]}
+{"filename": "image/image##hl_2405##test##17_00249.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##zc_zhugangchejian_frames##200036.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/image##dazuzhuang##0_0240.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "bbox": [538, 565, 713, 706], "is_ignored": false}, {"label": 1, "bbox": [0, 847, 123, 927], "is_ignored": false}]}
diff --git a/benchmark/ped_bev/image/image##dazuzhuang##0_0240.jpg b/benchmark/ped_bev/image/image##dazuzhuang##0_0240.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..aba4a303bf8c67ae0c749dd2524a633a7c489789
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang##0_0240.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang##0_0272.jpg b/benchmark/ped_bev/image/image##dazuzhuang##0_0272.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..580a0ce78832f3e084e6e855e08661057721ba0b
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang##0_0272.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang##11_0087.jpg b/benchmark/ped_bev/image/image##dazuzhuang##11_0087.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f22c17e9bbc7e05dc088d346ba0e4ae620a6a788
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang##11_0087.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang##3_0180.jpg b/benchmark/ped_bev/image/image##dazuzhuang##3_0180.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..aa3ca24d707ef438fe9b82046dda8ef8a9ef6e75
--- /dev/null
+++ b/benchmark/ped_bev/image/image##dazuzhuang##3_0180.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bff7a74e67f107f19a461db97ab0ece84f25db2b147385a0a87de239094a92c4
+size 100148
diff --git a/benchmark/ped_bev/image/image##dazuzhuang##8_0141.jpg b/benchmark/ped_bev/image/image##dazuzhuang##8_0141.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..d5c5cf0b0df8de8001c6cf8cea3090842e54e913
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang##8_0141.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang##9_0287.jpg b/benchmark/ped_bev/image/image##dazuzhuang##9_0287.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..be0630c40f4b317cef8b9f20080c5e37461dd062
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang##9_0287.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang##9_0356.jpg b/benchmark/ped_bev/image/image##dazuzhuang##9_0356.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..282ea30c25f5c4c77e9b9d88fd4bed851b8bca22
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang##9_0356.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_00480.jpg b/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_00480.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..7d9670ede29e7b2888bd7c93f8f57621c26c6bd9
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_00480.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_01200.jpg b/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_01200.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..1fa66d29d4173dbbfd3aea04885d4d39f04e6ae7
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_01200.jpg differ
diff --git a/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_01290.jpg b/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_01290.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3ca1b32bdb75dadc14062e255ed854f1f0dcbec6
Binary files /dev/null and b/benchmark/ped_bev/image/image##dazuzhuang_2##test##20_01290.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##11_00055.jpg b/benchmark/ped_bev/image/image##hl_2405##test##11_00055.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f4cf726e9b77ef0fcc34b0ee49d2f6ca44de2b2b
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##11_00055.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##12_00046.jpg b/benchmark/ped_bev/image/image##hl_2405##test##12_00046.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4af8968cb538a160e745b4cb0972c8bfd702e602
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##12_00046.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##13_00207.jpg b/benchmark/ped_bev/image/image##hl_2405##test##13_00207.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..159467976316efeb1bb07bca79087b010832d6a2
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##13_00207.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##17_00249.jpg b/benchmark/ped_bev/image/image##hl_2405##test##17_00249.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..72eb806753c295cca0746d81788b8f66ec0372e7
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##17_00249.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##18_00232.jpg b/benchmark/ped_bev/image/image##hl_2405##test##18_00232.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a729f202be33d06b1b8f9d354ce9138d9e1ea3c1
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##18_00232.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##3_00333.jpg b/benchmark/ped_bev/image/image##hl_2405##test##3_00333.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..2c1a980e56cddd2a0dcad8fe7a7401805dc50d96
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##3_00333.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##5_00048.jpg b/benchmark/ped_bev/image/image##hl_2405##test##5_00048.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..1532bb0f3e1e774d4ff7d078a7a4cb4aefcae2de
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##5_00048.jpg differ
diff --git a/benchmark/ped_bev/image/image##hl_2405##test##9_00076.jpg b/benchmark/ped_bev/image/image##hl_2405##test##9_00076.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..6df54a104664a92ba95b931469780e643ba58d24
Binary files /dev/null and b/benchmark/ped_bev/image/image##hl_2405##test##9_00076.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##100231.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##100231.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..02613a9705310234eb00e05f93bd430aedc7a58b
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##100231.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##200036.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##200036.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..7ff1aa9860ec2a5423b57d2dc7131c09763b7072
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##200036.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##200286.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##200286.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..77dbc416546d8cd26ff806087df58511255b0fdc
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##200286.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##300340.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##300340.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..204ffd04eea501a8847fccb56ac3a32de06188d7
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##300340.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##300471.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##300471.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..dfca11a8a88a70572a4b45b4059c2f307ad15219
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##300471.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##400520.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##400520.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4bb593ef08f4c41d3308a312ecd6de6f8d23da8f
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##400520.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500055.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500055.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..37b3e3ce6b52866891f22c3e7ce22fb932dd77be
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500055.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500144.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500144.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..fc21c24bd876892bcd1c697c1165861ebf800371
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500144.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500202.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500202.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..54cd49726e240f07ce53c1ff46e9526b1f361bc3
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500202.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500243.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500243.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..d27f9c6b06f87cd0825e77cce82983bb23c9b32c
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500243.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500283.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500283.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b0dc6884393dc39720a0221df1c8b4e47dd78ede
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500283.jpg differ
diff --git a/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500805.jpg b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500805.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..dfe703e64c9a16cdee9d244e6012e947a535dbc2
Binary files /dev/null and b/benchmark/ped_bev/image/image##zc_zhugangchejian_frames##500805.jpg differ
diff --git a/benchmark/ped_bev/image_description.jsonl b/benchmark/ped_bev/image_description.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..831470ed0182a1ac7237af63c827c89aaef88536
--- /dev/null
+++ b/benchmark/ped_bev/image_description.jsonl
@@ -0,0 +1,20 @@
+{"image_path": "image/image##dazuzhuang##8_0141.jpg", "image_description": "The image shows an overhead view of a factory floor with several workers and equipment. The timestamp on the image indicates it was taken on October 6, 2023, at 14:37:24. There are multiple workers wearing safety helmets and uniforms, engaged in various tasks. The floor is marked with yellow lines for safety and organization.\nIn the center of the image, there is a large piece of machinery being lifted by a crane. The machinery appears to be a complex assembly, possibly part of a larger industrial component. The workers around the machinery seem to be coordinating the lifting process or preparing the area for further work.\nOn the right side of the image, there is a blue structure that looks like a gantry crane or a similar type of lifting equipment. The floor has various other pieces of machinery and tools scattered around, indicating an active manufacturing environment.\nThere is no specific mention of \"ped bev\" in the image, but if you are referring to a particular piece of equipment or a worker's position, more details would be needed to identify it accurately."}
+{"image_path": "image/image##dazuzhuang##11_0087.jpg", "image_description": "The image shows an overhead view of a factory or industrial workspace. There are several workers wearing red hard hats and dark uniforms, indicating they are likely engaged in manual labor or manufacturing tasks. The floor is concrete, and there are various pieces of machinery and equipment scattered around, including what appears to be a large crane or lifting mechanism on the right side of the image.\nIn the bottom left corner, there is a vertical structure that could be part of the building's framework or a support column. The timestamp in the top right corner indicates the date and time as October 6, 2023, at 14:36:28.\nThere is no specific \"ped bev\" (pedestrian vehicle) visible in the image. The term \"ped bev\" typically refers to a pedestrian electric vehicle, which is not present in this industrial setting."}
+{"image_path": "image/image##zc_zhugangchejian_frames##300340.jpg", "image_description": "The image appears to be taken from a security camera inside an industrial facility, likely a foundry or metalworking plant. The timestamp on the image indicates it was captured on January 8, 2024, at 11:28:38. The environment is dimly lit with artificial lighting, and there is visible dust and debris, which is common in such industrial settings.\nIn the foreground, there is a large piece of machinery that looks like a casting machine or a part of a foundry setup. The machinery has a cylindrical component and several rectangular sections, possibly for holding molten metal during the casting process. The surrounding area includes metal structures and platforms, some of which have yellow safety railings.\nThere are no visible people in the image, but the presence of heavy machinery suggests that workers are likely operating the equipment out of view. The text overlay on the image provides information about the location, indicating it is related to \"Zongche Meishan Vehicles Co., Ltd.\" and mentions \"casting division\" and \"platform,\" suggesting this is a specific section within the company's premises."}
+{"image_path": "image/image##hl_2405##test##11_00055.jpg", "image_description": "The image shows an industrial setting, likely a foundry or metalworking facility, as indicated by the presence of heavy machinery and materials. The timestamp on the image reads \"2024-04-10 02:28:41,\" suggesting that the photo was taken at night.\nIn the center of the image, there is a large cylindrical object, possibly a mold or part of a casting process, being handled by a crane. The crane's hook is visible, and it appears to be lifting or moving the cylindrical object. There is a worker standing near the object, wearing dark clothing with a red patch on the back, indicating they might be part of the maintenance or operational staff.\nThe environment is dusty, which is typical for such industrial settings, and there are various pieces of equipment and machinery around, including what looks like a control panel and some pipes or conduits. The floor has yellow markings, which could indicate safety zones or designated areas within the facility.\nThere is no specific mention of a \"ped bev\" in the image, but if you are referring to a particular feature or object, it would need to be identified more clearly. If you have any additional details or context about what you are looking for, please provide them, and I can assist further."}
+{"image_path": "image/image##dazuzhuang##3_0180.jpg", "image_description": "The image shows an overhead view of a factory or industrial setting. There are several workers and pieces of machinery visible. The timestamp on the image indicates it was taken on October 6, 2023, at 15:38:01.\nIn the center of the image, there is a worker wearing a red hard hat and a gray shirt. This worker appears to be standing near a large piece of machinery that looks like a crane or lifting device. The worker seems to be inspecting or operating the machinery.\nTo the right of the central worker, there is another piece of machinery with a blue frame and a red control panel. A coiled hose is connected to this machinery, suggesting it might be used for some form of fluid transfer or cooling.\nOn the left side of the image, there are stacks of what appear to be metal sheets or panels. These are neatly arranged and secured with yellow straps.\nThe floor of the area is concrete, and there are various other pieces of equipment and tools scattered around, indicating an active work environment. The text in the bottom right corner of the image suggests that this is from a security camera feed, specifically from \"中车眉山车辆有限公司\" (China Railway Mian Shan Vehicle Co., Ltd.), which translates to \"China Railway Mian Shan Vehicle Company.\" The text also mentions \"货车制造中心\" (Truck Manufacturing Center) and \"站台\" (Platform), indicating the specific location within the company."}
+{"image_path": "image/image##dazuzhuang##9_0356.jpg", "image_description": "The image shows an overhead view of a factory or industrial setting. There is a large piece of machinery being lifted by a crane. The machinery appears to be a complex assembly with various components, including what looks like a black cylindrical part and some red structural elements. The floor has yellow and black striped markings, likely for safety and to indicate restricted areas.\nThere are two workers visible in the image. One worker is standing on the left side, wearing a yellow hard hat and dark clothing, while the other worker is partially visible on the right side, also wearing a hard hat and light-colored clothing. Both workers appear to be observing the lifting operation.\nThe timestamp in the top right corner indicates that the photo was taken on October 6, 2023, at 15:40:57. The text at the bottom of the image suggests that this is from a security camera at \"中车眉山车辆有限公司\" (China Railway Meishan Vehicle Co., Ltd.), specifically the \"货车制造中心\" (Freight Car Manufacturing Center) and \"站台\" (Platform)."}
+{"image_path": "image/image##dazuzhuang_2##test##20_01290.jpg", "image_description": "The image appears to be a security camera view from an industrial or manufacturing facility. The timestamp on the image indicates it was captured on April 23, 2024, at 15:56:44.\n### Description:\n- **Location**: The text in the bottom right corner suggests that this is located at \"中车眉山车辆有限公司\" (China Railway Mian Shan Vehicle Co., Ltd.), specifically within the \"货车制造中心\" (Freight Car Manufacturing Center) and \"站台\" (Platform).\n- **Environment**: The area looks like a workshop or assembly line with various equipment and materials scattered around.\n- **Objects**:\n  - There is a large white panel or structure in the center of the image, which could be part of a vehicle or machinery being assembled.\n  - Several cables and wires are visible, suggesting electrical or mechanical work.\n  - There are stacks of what appear to be metal sheets or panels on the left side.\n  - A person is partially visible near the bottom right, working on some equipment. This individual seems to be operating or inspecting something, possibly related to the machinery or assembly process.\n  - Various tools and equipment are scattered around the workspace, indicating an active work environment.\n### Target Ped BEV Identification:\n- **Target**: The person near the bottom right of the image appears to be the \"target ped BEV\" as per the instruction. However, without more context or specific criteria for identifying \"ped BEV,\" it's challenging to provide a detailed description beyond their general presence in the scene.\n- **Description**: The person is wearing dark clothing and appears to be engaged in some form of manual labor or inspection. They are standing close to a piece of equipment, possibly operating or adjusting it. Their exact role or activity cannot be determined from the image alone.\nIf you have any additional details or specific questions about the person or the environment, feel free to ask!"}
+{"image_path": "image/image##zc_zhugangchejian_frames##500243.jpg", "image_description": "The image shows an industrial setting, likely a foundry or a similar heavy manufacturing facility. The scene includes large machinery and equipment, with a focus on a crane lifting a bucket or container filled with what appears to be molten metal or some other material. The environment is dusty and industrial, with visible pipes, structures, and safety barriers.\nThere are no visible pedestrians or people in the immediate foreground, but there are two individuals wearing hard hats and safety gear in the background, possibly workers overseeing the operation. The timestamp in the top right corner indicates that the photo was taken on January 8, 2024, at 15:20:02.\nThe text at the bottom of the image suggests that this is from a surveillance camera, specifically labeled \"铸钢天车1\" (Zhuangang Tianche 1), which translates to \"Cast Steel Crane 1.\" This indicates that the camera is monitoring a specific crane used for handling cast steel materials. The location is identified as \"中车眉山车辆有限公司厂区铸钢分公司站台\" (Zhongtong Meishan Vehicle Co., Ltd. Factory Cast Steel Branch Platform), providing context about the specific area within the company where this activity is taking place."}
+{"image_path": "image/image##dazuzhuang_2##test##20_01200.jpg", "image_description": "The image shows an overhead view of a factory or industrial setting with several workers engaged in various tasks. The timestamp on the image indicates it was taken on April 23, 2024, at 15:55:14.\nIn the center of the image, there is a large white structure that appears to be part of a machine or assembly line. Several workers are around this structure:\n- One worker is standing near the top right corner of the white structure, possibly operating or inspecting something.\n- Another worker is positioned closer to the bottom left side of the structure, also appearing to work on or around it.\n- A third worker is located near the bottom right corner of the structure, possibly handling materials or equipment.\nThe floor is marked with various lines and areas, likely indicating different sections or pathways within the facility. There are also some boxes and other items scattered around the area, suggesting ongoing work or assembly processes.\nThe text in the bottom right corner of the image provides information about the location, which seems to be related to a company named \"中车眉山车辆有限公司\" (China Railway Mian Shan Vehicle Co., Ltd.), specifically mentioning a \"货车制造中心\" (Truck Manufacturing Center) and a \"站台\" (Platform). This suggests that the image might have been captured from a security camera monitoring the manufacturing activities in this specific area."}
+{"image_path": "image/image##dazuzhuang_2##test##20_00480.jpg", "image_description": "The image appears to be a security camera view from a factory or industrial setting. The timestamp on the image indicates it was taken on April 23, 2024, at 15:43:15. The location is identified as \"中车眉山车辆有限公司 厂区 货车制造中心 站台\" which translates to \"China Railway Meishan Vehicle Co., Ltd. Factory Freight Car Manufacturing Center Platform.\"\nIn the image, there is a large piece of machinery or equipment being worked on. It seems to be a part of a vehicle or a large industrial component, possibly a truck or train car, given the context of the manufacturing center. The equipment is covered with protective plastic and has various tools and cables around it.\nThere are two individuals visible in the bottom left corner of the image, both wearing safety helmets and working on the equipment. One person is standing while the other appears to be kneeling or crouching, possibly inspecting or working on a specific part of the machinery.\nThe environment includes a concrete floor, some cardboard boxes, and what looks like a red generator or power unit in the top left corner. There are also some cables and tools scattered around the area, indicating an active work site.\nNo specific target labeled as \"ped bev\" is present in the image. If you have any further details or need a more specific analysis, please let me know!"}
+{"image_path": "image/image##zc_zhugangchejian_frames##100231.jpg", "image_description": "The image appears to be a security camera view from a factory or industrial setting, specifically from a location labeled \"铸钢天车1\" (Cast Steel Crane 1). The timestamp on the image indicates it was taken on January 8, 2024, at 10:57:50.\nIn the center of the image, there is a large circular structure that looks like a casting mold or a part of a foundry process. A worker wearing blue protective clothing and a red hard hat is standing inside this circular structure, possibly inspecting or working on it. The environment is dusty, suggesting active work with molten metal or other materials.\nSurrounding the central structure are several cylindrical containers, likely used for storing raw materials or finished products. There are also various pieces of machinery and equipment scattered around the area, indicating an active manufacturing or processing facility.\nThe text in the bottom right corner provides additional context about the location, mentioning \"中车眉山车辆有限公司\" (China Railway Meishan Vehicle Co., Ltd.), \"厂区\" (Factory Area), \"铸钢分公司\" (Cast Steel Branch Company), and \"站台\" (Platform), which suggests that this is a specific section within the company dedicated to cast steel production."}
+{"image_path": "image/image##zc_zhugangchejian_frames##400520.jpg", "image_description": "The image appears to be a security camera view from a factory or industrial setting. The timestamp on the image indicates it was taken on January 8, 2024, at 11:48:39. The environment is dusty and industrial, with visible machinery and equipment. There are two individuals wearing safety gear, including hard hats, standing near a large piece of machinery that seems to be part of a casting or forging process. The machinery includes a yellow structure and various metal components.\nThere is no specific \"target ped bev\" mentioned in the image. If you have a specific object or area you're referring to as \"ped bev,\" please clarify, and I can provide a more detailed description."}
+{"image_path": "image/image##zc_zhugangchejian_frames##500805.jpg", "image_description": "The image appears to be taken from a security camera inside an industrial facility. The timestamp on the image indicates it was captured on January 8, 2024, at 15:29:25. The environment looks like a workshop or factory area with heavy machinery and equipment. There are visible signs of wear and dust, suggesting that the area is not regularly cleaned.\nIn the lower right corner of the image, there is text in Chinese characters. This text likely provides information about the location or the company associated with the footage. The text reads \"中车眉山车辆有限公司 厂区 铸钢分公司 站台,\" which translates to \"China Railway Meishan Vehicle Co., Ltd. Factory Cast Steel Branch Platform.\"\nThere is no specific target labeled as \"ped bev\" in the image. The focus seems to be on the general environment rather than any particular object or person. The image does not show any individuals or objects that could be identified as \"ped bev.\""}
+{"image_path": "image/image##hl_2405##test##13_00207.jpg", "image_description": "The image appears to be taken from a security camera at a manufacturing or industrial facility. The timestamp on the image indicates it was captured on April 10, 2024, at 3:10:22 AM. The environment looks like a workshop or factory floor with heavy machinery and equipment.\nIn the foreground, there is a large piece of machinery, possibly a crane or hoist, with cables hanging down. The floor is concrete, and there are various items scattered around, including what looks like metal sheets or plates. There is also a person wearing a red jacket and white helmet, who seems to be working near some equipment or machinery. The area is somewhat dusty, suggesting that work involving materials that generate dust is being performed.\nThe text in the bottom right corner of the image is in Chinese and provides information about the location, which appears to be related to a company named \"中车眉山车辆有限公司\" (China Railway Mian Shan Vehicle Co., Ltd.), specifically mentioning \"厂区铸钢分公司\" (Factory Casting Steel Branch) and \"站台\" (Platform). This suggests that the image is from a surveillance system monitoring an industrial casting operation."}
+{"image_path": "image/image##zc_zhugangchejian_frames##500144.jpg", "image_description": "The image appears to be a security camera view from a factory or industrial setting. The timestamp on the image indicates it was taken on January 8, 2024, at 15:18:21. The location is identified as \"中车眉山车辆有限公司\" (China Railway Meishan Locomotive Co., Ltd.), specifically the casting division.\nIn the image, there are two workers wearing blue uniforms and red hard hats. They are standing near a large piece of machinery, possibly a crane or lifting equipment, which is partially visible on the right side of the frame. The floor is concrete, and there are various industrial items scattered around, including what looks like a pile of cables or hoses and some metal structures.\nThe environment suggests an active work area with heavy machinery and industrial operations. There is no specific mention of a \"ped bev\" in the image, but the workers' presence and the machinery indicate that they might be involved in some form of industrial activity."}
+{"image_path": "image/image##zc_zhugangchejian_frames##500283.jpg", "image_description": "The image appears to be taken from a surveillance camera inside a factory or industrial setting. The timestamp on the image indicates it was captured on January 8, 2024, at 15:20:42.\nThe scene shows a large industrial area with various pieces of machinery and equipment. There are several cylindrical structures that resemble large tanks or vessels, possibly used for manufacturing processes such as casting or processing materials. The floor is marked with yellow lines, likely indicating safety boundaries or designated work areas.\nIn the foreground, there is a person wearing a red hard hat and dark clothing, standing near some machinery. The individual seems to be working or inspecting the area. The environment looks dusty, which could be due to the nature of the industrial processes taking place.\nThe text in the bottom right corner of the image is in Chinese and reads \"中车眉山车辆有限公司 厂区 铸钢分公司 站台,\" which translates to \"China Railway Meishan Vehicle Co., Ltd. Factory Casting Steel Branch Platform.\" This suggests that the location is part of a manufacturing facility specializing in casting steel products.\nThere is no visible target labeled as \"ped bev\" in the image. If you have any specific questions about the image or need further details, feel free to ask!"}
+{"image_path": "image/image##dazuzhuang##0_0272.jpg", "image_description": "The image appears to be a security camera view from an industrial setting, likely a factory or manufacturing plant. The timestamp on the image indicates it was taken on October 6, 2023, at 15:39:33.\nIn the image, there are several workers and pieces of machinery visible. The workers are wearing safety gear, including helmets and reflective vests, which suggests adherence to safety protocols in the workplace. The machinery includes large metal structures that appear to be part of the manufacturing process, possibly for assembling or transporting components.\nThere is no specific \"ped BEV\" (Personal Electric Vehicle) present in the image. The scene is focused on industrial activities rather than personal transportation. The environment looks like a busy work area with various equipment and materials scattered around."}
+{"image_path": "image/image##zc_zhugangchejian_frames##500202.jpg", "image_description": "The image appears to be a security camera view from an industrial setting, likely a factory or a construction site. The timestamp on the image indicates it was taken on January 8, 2024, at 15:19:21.\nIn the center of the image, there is a large piece of machinery, possibly a crane or a hoist, with cables extending downward. Below the machinery, there is a worker wearing a blue hard hat and safety gear, standing near what looks like a platform or a workbench. The worker seems to be operating or inspecting something on the ground.\nThe environment around the worker includes various industrial elements such as metal structures, platforms, and what appears to be a concrete floor. There are also some yellow safety railings visible, which are commonly used in industrial settings for safety purposes.\nThe text in the bottom right corner of the image suggests that the location might be related to \"Zhen Gang Tian Che 1,\" which could refer to a specific machine or area within the facility. The overall scene depicts a typical industrial workspace where heavy machinery and workers are involved in operations."}
+{"image_path": "image/image##zc_zhugangchejian_frames##300471.jpg", "image_description": "The image appears to be taken from a security camera inside an industrial facility, likely a foundry or metalworking plant. The timestamp on the image indicates it was captured on January 8, 2024, at 11:30:49 AM.\nIn the foreground, there is a large cylindrical object that looks like a casting mold or part of a machine used for casting metal. The surface of this object is rough and appears to be covered with some kind of residue or dust, which is common in such environments due to the high temperatures and materials involved in metal casting.\nTo the right of the cylindrical object, there is a person wearing a blue shirt and red pants, who seems to be working on or inspecting something near the floor. The individual is standing on a platform or a raised area, which is typical in industrial settings to allow workers to operate machinery safely.\nThe environment is dimly lit, with artificial lighting illuminating the workspace. There are various pieces of equipment and machinery visible in the background, including what looks like a crane or lifting mechanism, which is often used in foundries to move heavy objects.\nThe text overlay on the image includes Chinese characters, which translate to \"China Railway Mian Shan Locomotive Co., Ltd.\" and \"Casting Division Platform,\" indicating the location of the facility. The presence of this text suggests that the facility is related to the manufacturing of railway locomotives or similar heavy machinery."}
+{"image_path": "image/image##dazuzhuang##0_0240.jpg", "image_description": "The image appears to be a security camera view from a factory or industrial setting. The timestamp on the image indicates it was taken on October 6, 2023, at 15:39:01. The location is identified as \"中车眉山车辆有限公司 厂区 货车制造中心 站台\" which translates to \"China Railway Mian Shan Vehicle Co., Ltd. Factory Freight Car Manufacturing Center Platform.\"\nIn the image, there are two red rectangular structures that resemble platforms or workstations. A person wearing a white shirt and blue pants is standing on one of these platforms. There is also another individual wearing a red shirt and dark pants lying on the ground near the edge of the platform. The person in the red shirt seems to be in a prone position.\nThe area is equipped with various industrial equipment, including cranes and cables hanging from the ceiling. The floor is concrete, and there are some yellow safety markings visible. The overall environment suggests an active industrial workspace."}
diff --git a/benchmark/ped_bev/user_input.txt b/benchmark/ped_bev/user_input.txt
new file mode 100644
index 0000000000000000000000000000000000000000..f42022a40b72c6568e9d4ad74ddc8a6495e90733
--- /dev/null
+++ b/benchmark/ped_bev/user_input.txt
@@ -0,0 +1 @@
+检测车间俯拍的工人
\ No newline at end of file
diff --git "a/benchmark/ped_bev/\345\217\202\350\200\203\345\233\276\347\211\207/image##dazuzhuang##0_0240.jpg" "b/benchmark/ped_bev/\345\217\202\350\200\203\345\233\276\347\211\207/image##dazuzhuang##0_0240.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..aba4a303bf8c67ae0c749dd2524a633a7c489789
Binary files /dev/null and "b/benchmark/ped_bev/\345\217\202\350\200\203\345\233\276\347\211\207/image##dazuzhuang##0_0240.jpg" differ
diff --git a/benchmark/smog/gt.jsonl b/benchmark/smog/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..9189c0b1598d836426dd40ba28cab645923871f9
--- /dev/null
+++ b/benchmark/smog/gt.jsonl
@@ -0,0 +1,30 @@
+{"filename": "image/iqef9gA6ALnvT3l1DDreo.jpg", "image_height": 562, "image_width": 1000, "instances": [{"label": 1, "label_name": "smog", "bbox": [5, 9, 542, 411]}]}
+{"filename": "image/BGP8kQvB9XhKZeHfRoIkL.jpeg", "image_height": 810, "image_width": 1080, "instances": [{"label": 1, "label_name": "smog", "bbox": [915, 472, 1073, 598]}, {"label": 1, "label_name": "smog", "bbox": [699, 410, 951, 561]}]}
+{"filename": "image/PU-hwNNBmEC12dZB7vEGJ.jpg", "image_height": 690, "image_width": 690, "instances": [{"label": 1, "label_name": "smog", "bbox": [300, 215, 680, 561]}, {"label": 1, "label_name": "smog", "bbox": [58, 125, 138, 235]}]}
+{"filename": "image/X1_NbELlrcKd7PUkoeEsi.jpg", "image_height": 853, "image_width": 480, "instances": [{"label": 1, "label_name": "smog", "bbox": [142, 1, 474, 452]}]}
+{"filename": "image/wizgEUAJ5GulK13NITOu7.jpeg", "image_height": 2208, "image_width": 1242, "instances": []}
+{"filename": "image/-xGxHXYw3G6VvF0X6LHzM.jpg", "image_height": 1440, "image_width": 1080, "instances": [{"label": 1, "label_name": "smog", "bbox": [15, 394, 483, 683]}, {"label": 1, "label_name": "smog", "bbox": [390, 223, 781, 519]}, {"label": 1, "label_name": "smog", "bbox": [885, 257, 1072, 419]}, {"label": 1, "label_name": "smog", "bbox": [860, 11, 1037, 223]}]}
+{"filename": "image/tuWVK9G0eu9dacLrElQbD.jpg", "image_height": 1080, "image_width": 1920, "instances": []}
+{"filename": "image/yZBBxHryxhMEfFX5zKMWs.jpg", "image_height": 720, "image_width": 544, "instances": [{"label": 1, "label_name": "smog", "bbox": [224, 133, 321, 281]}]}
+{"filename": "image/_rXwIsIn23KpX_NIZU8fC.jpg", "image_height": 720, "image_width": 740, "instances": [{"label": 1, "label_name": "smog", "bbox": [69, 233, 310, 370]}]}
+{"filename": "image/yDDmmcEzFjJG62qvVh8oh.jpg", "image_height": 1280, "image_width": 2263, "instances": [{"label": 1, "label_name": "smog", "bbox": [680, 268, 1526, 1039]}]}
+{"filename": "image/lJqnoPxfepG7vPyL8wcEE.jpg", "image_height": 1124, "image_width": 1500, "instances": [{"label": 1, "label_name": "smog", "bbox": [387, 779, 498, 847]}, {"label": 1, "label_name": "smog", "bbox": [559, 606, 703, 729]}, {"label": 1, "label_name": "smog", "bbox": [715, 536, 1039, 703]}]}
+{"filename": "image/s_beSQsdI4H2ZKzINRYFo.jpg", "image_height": 414, "image_width": 228, "instances": [{"label": 1, "label_name": "smog", "bbox": [83, 4, 224, 260]}]}
+{"filename": "image/_bmGA7_LmrEOWFCcW0POp.jpg", "image_height": 412, "image_width": 228, "instances": [{"label": 1, "label_name": "smog", "bbox": [34, 167, 64, 196]}, {"label": 1, "label_name": "smog", "bbox": [91, 80, 223, 293]}]}
+{"filename": "image/QDNLLw2rz2zd7OqYq-XsL.jpeg", "image_height": 494, "image_width": 640, "instances": [{"label": 1, "label_name": "smog", "bbox": [240, 213, 316, 328]}, {"label": 1, "label_name": "smog", "bbox": [339, 27, 472, 289]}]}
+{"filename": "image/liBRM7GLZTTnNr43skYCk.jpg", "image_height": 1034, "image_width": 800, "instances": [{"label": 1, "label_name": "smog", "bbox": [278, 527, 357, 618]}, {"label": 1, "label_name": "smog", "bbox": [285, 169, 551, 589]}]}
+{"filename": "image/dbkdbFKii3AY1WB9Ddn3g.jpg", "image_height": 1280, "image_width": 2327, "instances": []}
+{"filename": "image/km8YIqcvHVZTvZm2BPzDN.jpeg", "image_height": 1920, "image_width": 1080, "instances": [{"label": 1, "label_name": "smog", "bbox": [559, 15, 778, 807]}]}
+{"filename": "image/r4wgU8UN8zvdxgOzCDJrh.jpg", "image_height": 968, "image_width": 1500, "instances": [{"label": 1, "label_name": "smog", "bbox": [23, 286, 406, 629]}, {"label": 1, "label_name": "smog", "bbox": [369, 372, 1477, 960]}]}
+{"filename": "image/x_2joRmo8KHRQfveORsMh.jpg", "image_height": 1915, "image_width": 1080, "instances": [{"label": 1, "label_name": "smog", "bbox": [652, 904, 1062, 1083]}]}
+{"filename": "image/Va4djcAIDTBHxkEWh3sR9.jpeg", "image_height": 1024, "image_width": 680, "instances": [{"label": 1, "label_name": "smog", "bbox": [362, 201, 672, 546]}]}
+{"filename": "image/Mlge558eDQglMeFC2d-tT.jpeg", "image_height": 1067, "image_width": 800, "instances": [{"label": 1, "label_name": "smog", "bbox": [298, 310, 381, 474]}, {"label": 1, "label_name": "smog", "bbox": [411, 323, 456, 471]}]}
+{"filename": "image/ua_VWNMQkwRFILd4wgeIh.jpg", "image_height": 333, "image_width": 500, "instances": [{"label": 1, "label_name": "smog", "bbox": [35, 76, 199, 169]}]}
+{"filename": "image/cX0RIMVZZuceLikcep7zU.jpeg", "image_height": 480, "image_width": 640, "instances": [{"label": 1, "label_name": "smog", "bbox": [336, 2, 426, 210]}]}
+{"filename": "image/ZgOaHSLAgWggjuMS9GPHI.jpeg", "image_height": 1067, "image_width": 800, "instances": []}
+{"filename": "image/DhvK2RDK6-G9WaLvhqYhe.jpg", "image_height": 667, "image_width": 500, "instances": [{"label": 1, "label_name": "smog", "bbox": [132, 103, 324, 512]}]}
+{"filename": "image/onneNwxD9igAEwCt1zwOV.jpeg", "image_height": 667, "image_width": 1200, "instances": [{"label": 1, "label_name": "smog", "bbox": [478, 217, 573, 291]}, {"label": 1, "label_name": "smog", "bbox": [588, 211, 632, 290]}]}
+{"filename": "image/-NJ5xBiM8i-ThNAVKGbYG.jpeg", "image_height": 1067, "image_width": 800, "instances": [{"label": 1, "label_name": "smog", "bbox": [245, 170, 365, 594]}, {"label": 1, "label_name": "smog", "bbox": [328, 192, 448, 618]}, {"label": 1, "label_name": "smog", "bbox": [519, 203, 617, 620]}]}
+{"filename": "image/4mBy6LRkPucdhu7XxgEot.jpg", "image_height": 500, "image_width": 889, "instances": [{"label": 1, "label_name": "smog", "bbox": [9, 143, 173, 256]}, {"label": 1, "label_name": "smog", "bbox": [271, 141, 340, 189]}, {"label": 1, "label_name": "smog", "bbox": [403, 81, 553, 259]}, {"label": 1, "label_name": "smog", "bbox": [525, 206, 673, 493]}]}
+{"filename": "image/vWF4bdevHe3GqcGtz8Iwv.jpg", "image_height": 426, "image_width": 640, "instances": [{"label": 1, "label_name": "smog", "bbox": [108, 198, 197, 232]}, {"label": 1, "label_name": "smog", "bbox": [204, 204, 265, 234]}]}
+{"filename": "image/t09pz0CWWNaOTkC1B64SK.jpg", "image_height": 400, "image_width": 714, "instances": [{"label": 1, "label_name": "smog", "bbox": [202, 1, 403, 133]}, {"label": 1, "label_name": "smog", "bbox": [607, 186, 644, 240]}, {"label": 1, "label_name": "smog", "bbox": [535, 191, 582, 225]}]}
diff --git a/benchmark/smog/image/-NJ5xBiM8i-ThNAVKGbYG.jpeg b/benchmark/smog/image/-NJ5xBiM8i-ThNAVKGbYG.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..49e89b4a753bee184f643c5551bebf78e6ce655e
Binary files /dev/null and b/benchmark/smog/image/-NJ5xBiM8i-ThNAVKGbYG.jpeg differ
diff --git a/benchmark/smog/image/-xGxHXYw3G6VvF0X6LHzM.jpg b/benchmark/smog/image/-xGxHXYw3G6VvF0X6LHzM.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..d303dd9ba61b827e243e3960718120a000e24bef
--- /dev/null
+++ b/benchmark/smog/image/-xGxHXYw3G6VvF0X6LHzM.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b07b8a3965698e84ad383d9ef596bf26fc91b6af325f595cb4c133b621278a8
+size 388868
diff --git a/benchmark/smog/image/4mBy6LRkPucdhu7XxgEot.jpg b/benchmark/smog/image/4mBy6LRkPucdhu7XxgEot.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..63ac24c51da0f6c19707feca62f0364eb3f491d1
Binary files /dev/null and b/benchmark/smog/image/4mBy6LRkPucdhu7XxgEot.jpg differ
diff --git a/benchmark/smog/image/BGP8kQvB9XhKZeHfRoIkL.jpeg b/benchmark/smog/image/BGP8kQvB9XhKZeHfRoIkL.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..caa8aae64d57aaa53f4ead3b808bb29d647ed916
--- /dev/null
+++ b/benchmark/smog/image/BGP8kQvB9XhKZeHfRoIkL.jpeg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a4dd1645ba6a16c662251435df5faf90b4b533c146337a8ad06c2084e3d34c6
+size 132119
diff --git a/benchmark/smog/image/DhvK2RDK6-G9WaLvhqYhe.jpg b/benchmark/smog/image/DhvK2RDK6-G9WaLvhqYhe.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f94ff68f38db76a209b25a257e61d8f7f8477c4f
Binary files /dev/null and b/benchmark/smog/image/DhvK2RDK6-G9WaLvhqYhe.jpg differ
diff --git a/benchmark/smog/image/Mlge558eDQglMeFC2d-tT.jpeg b/benchmark/smog/image/Mlge558eDQglMeFC2d-tT.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..afed6e6f17897cf6e7e96b87e0fed0f0b2477534
Binary files /dev/null and b/benchmark/smog/image/Mlge558eDQglMeFC2d-tT.jpeg differ
diff --git a/benchmark/smog/image/PU-hwNNBmEC12dZB7vEGJ.jpg b/benchmark/smog/image/PU-hwNNBmEC12dZB7vEGJ.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..963857d8b461dd51430afb3170aa86243bb6de8f
Binary files /dev/null and b/benchmark/smog/image/PU-hwNNBmEC12dZB7vEGJ.jpg differ
diff --git a/benchmark/smog/image/QDNLLw2rz2zd7OqYq-XsL.jpeg b/benchmark/smog/image/QDNLLw2rz2zd7OqYq-XsL.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..830269924ca40633e6280c159c5a5ef9e2b492c4
Binary files /dev/null and b/benchmark/smog/image/QDNLLw2rz2zd7OqYq-XsL.jpeg differ
diff --git a/benchmark/smog/image/Va4djcAIDTBHxkEWh3sR9.jpeg b/benchmark/smog/image/Va4djcAIDTBHxkEWh3sR9.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..be54d0257cedb19bbda45d97047128529071aa3c
--- /dev/null
+++ b/benchmark/smog/image/Va4djcAIDTBHxkEWh3sR9.jpeg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8737b78f6f479dc4e255a1bd69154a3f1ed05fe4ceb375966f6944613da25e7c
+size 115441
diff --git a/benchmark/smog/image/X1_NbELlrcKd7PUkoeEsi.jpg b/benchmark/smog/image/X1_NbELlrcKd7PUkoeEsi.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..5482b5688c6be62246c2acc59a0de14543cf0455
Binary files /dev/null and b/benchmark/smog/image/X1_NbELlrcKd7PUkoeEsi.jpg differ
diff --git a/benchmark/smog/image/ZgOaHSLAgWggjuMS9GPHI.jpeg b/benchmark/smog/image/ZgOaHSLAgWggjuMS9GPHI.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..c36cb591efce57c8736f84bac2127126fa13dfb1
--- /dev/null
+++ b/benchmark/smog/image/ZgOaHSLAgWggjuMS9GPHI.jpeg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cba2d8c006ce262056a764638e97f4128e8d6237ee07b1461666ba9d7e24685f
+size 131728
diff --git a/benchmark/smog/image/_bmGA7_LmrEOWFCcW0POp.jpg b/benchmark/smog/image/_bmGA7_LmrEOWFCcW0POp.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..9e7616f47697f0c7d8b1229a93a9f3c692690888
Binary files /dev/null and b/benchmark/smog/image/_bmGA7_LmrEOWFCcW0POp.jpg differ
diff --git a/benchmark/smog/image/_rXwIsIn23KpX_NIZU8fC.jpg b/benchmark/smog/image/_rXwIsIn23KpX_NIZU8fC.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..eb07f2ec132c42d992983d3ec59fd9ddc5389c04
Binary files /dev/null and b/benchmark/smog/image/_rXwIsIn23KpX_NIZU8fC.jpg differ
diff --git a/benchmark/smog/image/cX0RIMVZZuceLikcep7zU.jpeg b/benchmark/smog/image/cX0RIMVZZuceLikcep7zU.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..00798528cf0bf98a94edc5381f00a5954d764054
Binary files /dev/null and b/benchmark/smog/image/cX0RIMVZZuceLikcep7zU.jpeg differ
diff --git a/benchmark/smog/image/dbkdbFKii3AY1WB9Ddn3g.jpg b/benchmark/smog/image/dbkdbFKii3AY1WB9Ddn3g.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..6e32a653605fe889280234ad44bc4ada4223ce12
--- /dev/null
+++ b/benchmark/smog/image/dbkdbFKii3AY1WB9Ddn3g.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cf3e631d4e259cf79514ccd3641a0adf56225d936877729360819980e1c5a6a
+size 613741
diff --git a/benchmark/smog/image/iqef9gA6ALnvT3l1DDreo.jpg b/benchmark/smog/image/iqef9gA6ALnvT3l1DDreo.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..7dc7d53602ab7fd851e009050b881bf5d4a65ace
Binary files /dev/null and b/benchmark/smog/image/iqef9gA6ALnvT3l1DDreo.jpg differ
diff --git a/benchmark/smog/image/km8YIqcvHVZTvZm2BPzDN.jpeg b/benchmark/smog/image/km8YIqcvHVZTvZm2BPzDN.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..e1173a3cf1fe976a61ca031019cabe2d2566a908
--- /dev/null
+++ b/benchmark/smog/image/km8YIqcvHVZTvZm2BPzDN.jpeg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ba4541a509c50175bb79189754ae411486ec3b2b45e16283556610fc9f5686e
+size 253926
diff --git a/benchmark/smog/image/lJqnoPxfepG7vPyL8wcEE.jpg b/benchmark/smog/image/lJqnoPxfepG7vPyL8wcEE.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..ba11c0ee0770d6a2f9ca19c52de890ae644abfac
--- /dev/null
+++ b/benchmark/smog/image/lJqnoPxfepG7vPyL8wcEE.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d8679b0d4c0365641933b3887b894ac448837c8f3ad42012466d828e2aedd53
+size 642806
diff --git a/benchmark/smog/image/liBRM7GLZTTnNr43skYCk.jpg b/benchmark/smog/image/liBRM7GLZTTnNr43skYCk.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e9c287f931c9e596813fc74891da09caa5ad6c8c
--- /dev/null
+++ b/benchmark/smog/image/liBRM7GLZTTnNr43skYCk.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:150a8d7f5092eda356fdadb1ad529d60df4d4a71af9a6fe8f9d143eded922aac
+size 143250
diff --git a/benchmark/smog/image/onneNwxD9igAEwCt1zwOV.jpeg b/benchmark/smog/image/onneNwxD9igAEwCt1zwOV.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..9718aa0d5f9f494a1e1c6b104e2342865dbfb4d7
Binary files /dev/null and b/benchmark/smog/image/onneNwxD9igAEwCt1zwOV.jpeg differ
diff --git a/benchmark/smog/image/r4wgU8UN8zvdxgOzCDJrh.jpg b/benchmark/smog/image/r4wgU8UN8zvdxgOzCDJrh.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..9565c62ee98f947c177aa5ed6273532413e9cdfd
--- /dev/null
+++ b/benchmark/smog/image/r4wgU8UN8zvdxgOzCDJrh.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eaeffd96d716d52b2e7f68c150156468219c78ebf964b360b221aabca9c301e0
+size 470406
diff --git a/benchmark/smog/image/s_beSQsdI4H2ZKzINRYFo.jpg b/benchmark/smog/image/s_beSQsdI4H2ZKzINRYFo.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..9c5bb3513b34d9784f481156361ccf8f0d959ee5
Binary files /dev/null and b/benchmark/smog/image/s_beSQsdI4H2ZKzINRYFo.jpg differ
diff --git a/benchmark/smog/image/t09pz0CWWNaOTkC1B64SK.jpg b/benchmark/smog/image/t09pz0CWWNaOTkC1B64SK.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..c2a70ad2fb7b5ea8cc341db064e9b85c8ad5f18e
Binary files /dev/null and b/benchmark/smog/image/t09pz0CWWNaOTkC1B64SK.jpg differ
diff --git a/benchmark/smog/image/tuWVK9G0eu9dacLrElQbD.jpg b/benchmark/smog/image/tuWVK9G0eu9dacLrElQbD.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..c6910c8372114b664015eef5f44d07d6125867f9
--- /dev/null
+++ b/benchmark/smog/image/tuWVK9G0eu9dacLrElQbD.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2feec307e629686a59b78ba570a34a106669f5fdc8dde73d85b06db5235c4071
+size 269644
diff --git a/benchmark/smog/image/ua_VWNMQkwRFILd4wgeIh.jpg b/benchmark/smog/image/ua_VWNMQkwRFILd4wgeIh.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..60ba62ca9ff0a4f3366775b11c3899ada912fdae
Binary files /dev/null and b/benchmark/smog/image/ua_VWNMQkwRFILd4wgeIh.jpg differ
diff --git a/benchmark/smog/image/vWF4bdevHe3GqcGtz8Iwv.jpg b/benchmark/smog/image/vWF4bdevHe3GqcGtz8Iwv.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..aefcd95762bf9e170da151c0e2f04f23e2006eff
Binary files /dev/null and b/benchmark/smog/image/vWF4bdevHe3GqcGtz8Iwv.jpg differ
diff --git a/benchmark/smog/image/wizgEUAJ5GulK13NITOu7.jpeg b/benchmark/smog/image/wizgEUAJ5GulK13NITOu7.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..232cb2d7e7a1afd925997c6fe45e3e994e558850
--- /dev/null
+++ b/benchmark/smog/image/wizgEUAJ5GulK13NITOu7.jpeg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c4ec44ceac8b024b6ab54c3badd27df296e43506f2612f3e6afea07923238eb
+size 1947022
diff --git a/benchmark/smog/image/x_2joRmo8KHRQfveORsMh.jpg b/benchmark/smog/image/x_2joRmo8KHRQfveORsMh.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..673a875d5f5aacc9fd3e8ab7a1d47354af0105d8
--- /dev/null
+++ b/benchmark/smog/image/x_2joRmo8KHRQfveORsMh.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d88ab9bd96d30cbf0e9c1fb845471259cbd123e74253ae1609be167a4a513a22
+size 1151507
diff --git a/benchmark/smog/image/yDDmmcEzFjJG62qvVh8oh.jpg b/benchmark/smog/image/yDDmmcEzFjJG62qvVh8oh.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..f5f0c1768478ac744180b31874444feab095b66e
--- /dev/null
+++ b/benchmark/smog/image/yDDmmcEzFjJG62qvVh8oh.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6330af1274588dfd8feb8da499b53929ac3c5d530d3e8eeb26c33b62422bce0e
+size 655013
diff --git a/benchmark/smog/image/yZBBxHryxhMEfFX5zKMWs.jpg b/benchmark/smog/image/yZBBxHryxhMEfFX5zKMWs.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..fa37062dd2fb9fd3227bd921d79b48bdb25f7b0f
Binary files /dev/null and b/benchmark/smog/image/yZBBxHryxhMEfFX5zKMWs.jpg differ
diff --git a/benchmark/smog/image_description.jsonl b/benchmark/smog/image_description.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..219e15d74039816e8e95fa44eea8f9a8a40cd0f3
--- /dev/null
+++ b/benchmark/smog/image_description.jsonl
@@ -0,0 +1,26 @@
+{"image_path": "image/iqef9gA6ALnvT3l1DDreo.jpg", "image_description": "The image shows an aerial view of a landscape with a significant amount of white smoke or vapor rising from a specific area. The smoke appears to be concentrated and is billowing upwards, indicating some form of activity that is producing it. The surrounding area includes patches of greenery, possibly trees and shrubs, and what looks like a dirt or gravel path. There are also some structures visible, including what might be buildings or parts of a facility, but they are not clearly defined due to the distance and the presence of the smoke. The overall scene suggests an event or incident that is causing the release of the smoke."}
+{"image_path": "image/BGP8kQvB9XhKZeHfRoIkL.jpeg", "image_description": "The image shows an outdoor scene with several individuals wearing uniforms and helmets, likely firefighters or emergency responders, working near a pile of debris and blue corrugated metal sheets. The area appears to be a field or open space with some buildings in the background. There is visible smoke rising from the debris, indicating that there might be a fire or a controlled burn. The sky is clear and blue, suggesting it is daytime.\nRegarding the presence of smog, the image does not show any significant haze or pollution that would typically be associated with smog. The smoke seen in the image is localized and appears to be from the burning debris, which is a different phenomenon from smog. Smog usually refers to a more widespread, persistent haze in the atmosphere caused by air pollution, which is not evident in this image."}
+{"image_path": "image/PU-hwNNBmEC12dZB7vEGJ.jpg", "image_description": "The image depicts a traditional village with white-washed buildings and dark tiled roofs. The scene is bathed in a soft, golden light, suggesting either early morning or late afternoon. There is a noticeable plume of smoke rising from one of the buildings, indicating that something is being cooked or heated inside. The smoke is thick and billows upwards, creating a contrast against the otherwise serene and quiet atmosphere of the village. The surrounding area includes lush greenery, adding to the tranquil rural setting."}
+{"image_path": "image/X1_NbELlrcKd7PUkoeEsi.jpg", "image_description": "The image shows a scene at an airport with a significant amount of black smoke rising into the sky. The smoke appears to be coming from a distant location, possibly indicating a fire or some other incident. In the foreground, there are airplanes parked on the tarmac, including one with the logo \"Capital Airlines.\" The text overlay in Chinese provides information about a fire that occurred in Guangzhou Huadu District, near Baiyun Airport, and mentions that the fire has been extinguished with no casualties reported. The text also indicates that firefighters have been dispatched to the scene."}
+{"image_path": "image/-xGxHXYw3G6VvF0X6LHzM.jpg", "image_description": "The image shows a fire burning in an outdoor area, likely a field or a rural setting. The fire is consuming dry grass and vegetation, with flames visible and smoke rising into the air. There are utility poles and wires in the background, indicating that the location might be near a road or a power line. The sky appears overcast, and the overall atmosphere suggests a cloudy day.\nRegarding the presence of smog, there is indeed a significant amount of smoke visible in the image. The smoke is thick and billowing, extending from the fire and spreading across the scene. It obscures parts of the background and adds to the overall hazy appearance of the image. The smoke appears to be dense and dark, indicating that it could be harmful to respiratory health if inhaled."}
+{"image_path": "image/yZBBxHryxhMEfFX5zKMWs.jpg", "image_description": "The image depicts a winter scene with a factory or industrial building emitting smoke from its chimney. The sun is low on the horizon, casting a warm glow and creating a lens flare effect. The foreground shows a snow-covered field with tire tracks, suggesting recent vehicle activity. The background includes leafless trees, indicating it is likely late autumn or winter. The smoke from the chimney appears as a plume rising into the sky, adding to the industrial atmosphere of the scene.\n### Target Smog Description:\nThe target smog in the image is the visible plume of smoke emanating from the factory's chimney. It is a light gray color and contrasts against the clear blue sky. The smoke is thick and dense at the base of the chimney, gradually thinning out as it rises higher into the air. This plume is the primary focus of the image, highlighting the industrial activity and potentially the environmental impact of the emissions."}
+{"image_path": "image/_rXwIsIn23KpX_NIZU8fC.jpg", "image_description": "The image depicts a serene sunset scene with a gradient sky transitioning from blue to orange and yellow near the horizon. The sun is partially visible, casting a warm glow over the landscape. In the foreground, there is a body of water reflecting the colors of the sky, surrounded by silhouettes of trees and grasses. On the left side of the image, there is an industrial structure emitting dark smoke into the air, which appears as a plume rising against the backdrop of the setting sun. The smoke is thick and contrasts sharply with the lighter colors of the sky, indicating pollution or emissions from the industrial facility."}
+{"image_path": "image/yDDmmcEzFjJG62qvVh8oh.jpg", "image_description": "The image shows an outdoor scene with a large plume of smoke rising into the sky. The smoke appears to be billowing from a distant location, possibly indicating an explosion or some form of industrial activity. In the foreground, there are buildings and trees, suggesting an urban or semi-urban environment. On the right side of the image, there is a tall structure that resembles a tower or a chimney, which could be related to the source of the smoke.\nThe text \"ВОЕННАЯ ХРОНИКА\" (Military Chronicle) is visible in the top left corner, indicating that the image might be related to a military or conflict-related event. Additionally, there is Chinese text on the right side of the image, which translates to \"Depth Workshop,\" possibly referring to a media outlet or a segment title. The overall scene suggests a significant event, but without more context, it's difficult to determine the exact nature of the incident."}
+{"image_path": "image/lJqnoPxfepG7vPyL8wcEE.jpg", "image_description": "The image shows an aerial view of a large industrial complex, likely a chemical or petrochemical plant, with numerous interconnected structures and pipes. There is visible smoke or steam rising from one of the chimneys, indicating active operations within the facility. The surrounding area includes roads, some greenery, and residential buildings in the distance. The sky appears hazy, suggesting the presence of smog, which is a mixture of air pollutants that can reduce visibility and quality of life.\n### Description of the Target Smog:\n- **Location**: The smog is located near the chimney on the right side of the image.\n- **Appearance**: It is a white, billowing cloud of smoke or steam, which is typical of industrial emissions.\n- **Extent**: The plume extends horizontally for a short distance before dispersing into the atmosphere.\n- **Impact**: This type of emission contributes to air pollution, which can have adverse effects on human health and the environment.\nThe overall scene highlights the environmental impact of industrial activities, particularly those involving the release of pollutants into the air."}
+{"image_path": "image/s_beSQsdI4H2ZKzINRYFo.jpg", "image_description": "The image shows an industrial area with a tall chimney emitting a plume of white smoke or steam into the sky. The surrounding area includes residential buildings with red roofs and some commercial structures. There are also parked cars visible near the buildings. In the background, there is a mountain covered with trees, and the sky appears to be transitioning from day to evening, with a soft gradient of colors.\nThe target \"smog\" in this context refers to the white plume of smoke or steam coming from the chimney. It is thick and billowing, indicating that it is likely a significant amount of emissions being released into the atmosphere. The smoke appears to be rising high above the surrounding buildings, suggesting it could have an impact on air quality in the vicinity."}
+{"image_path": "image/_bmGA7_LmrEOWFCcW0POp.jpg", "image_description": "The image shows an urban landscape with a prominent smokestack emitting a large plume of white smoke into the air. The smoke appears to be thick and billowing, indicating significant emissions. The background features a mix of residential and commercial buildings, with some greenery visible on the hills in the distance. The sky is overcast, which might contribute to the visibility of the smoke.\n### Target Smog Description:\n- **Location**: The target smog is located directly above the smokestack, where the white smoke is most concentrated.\n- **Appearance**: The smoke is dense and opaque, suggesting it contains a high concentration of particulate matter or other pollutants.\n- **Extent**: The smoke extends horizontally for several hundred meters, indicating a significant emission source.\n- **Color**: The smoke is predominantly white, which could be due to the presence of water vapor or other compounds in the exhaust.\nThis image likely represents an industrial or power generation facility contributing to air pollution in the area."}
+{"image_path": "image/QDNLLw2rz2zd7OqYq-XsL.jpeg", "image_description": "The image shows three individuals working together to control a fire in a field. The fire appears to be burning dry grass and vegetation. The area is surrounded by green fields, and there are power lines running across the background. The sky is overcast, suggesting cloudy weather.\nRegarding the presence of smog, the image does not show any visible smog. However, the smoke from the fire can be seen rising into the air, which might contribute to some haze or reduced visibility in the immediate vicinity. The overall scene suggests an effort to manage a controlled burn rather than an uncontrolled wildfire."}
+{"image_path": "image/liBRM7GLZTTnNr43skYCk.jpg", "image_description": "The image depicts a scene on a river with numerous boats docked along the shore. The boats appear to be traditional fishing vessels, some covered with tarps and others with bamboo structures. There is a narrow path or walkway between the boats where people are walking and standing. The atmosphere is lively, with smoke rising from some areas, suggesting that there might be some form of cooking or burning activity taking place.\nIn the center of the image, there is a couple dressed in formal attire, possibly celebrating a wedding, as indicated by their attire and the presence of red decorations, which are often associated with weddings in many cultures. The couple is surrounded by other people who seem to be part of the celebration.\nThe water appears calm, and the overall setting suggests a community event or festival taking place on the river. The presence of smoke indicates that there might be food being cooked or offerings being made, which is common in many cultural celebrations."}
+{"image_path": "image/km8YIqcvHVZTvZm2BPzDN.jpeg", "image_description": "The image shows a tall residential building with a yellowish facade. There is a column of dark smoke rising from the top of the building, indicating a fire or some other incident. The sky is clear and blue, suggesting that the smoke is the primary focus. In the foreground, there is a glass-covered bus stop with some trees and shrubs around it. The overall scene appears to be an urban area during daylight hours."}
+{"image_path": "image/r4wgU8UN8zvdxgOzCDJrh.jpg", "image_description": "The image depicts a nighttime aerial view of a cityscape, likely a major urban center, with numerous illuminated skyscrapers and buildings. The scene is partially obscured by dense fog or low-hanging clouds, which create a dramatic and ethereal atmosphere. The lights from the buildings cast a colorful glow, with hues of blue, orange, and white dominating the scene. The fog appears to be thick enough to obscure parts of the city below, adding a sense of mystery and grandeur to the image.\nThe target \"smog\" in the image refers to the dense fog or low-hanging clouds that dominate the lower portion of the image. This smog is not a physical pollutant but rather a natural phenomenon that adds to the visual impact of the scene. It creates a layer that separates the viewer from the city below, enhancing the sense of depth and scale. The smog is thick and opaque, with some areas appearing lighter due to the interplay of light and shadow, while other areas are darker, indicating varying thicknesses of the fog."}
+{"image_path": "image/x_2joRmo8KHRQfveORsMh.jpg", "image_description": "The image depicts a serene scene at sunset or sunrise over a body of water. The sky is a gradient of colors, transitioning from a soft blue at the top to warm hues of orange and yellow near the horizon where the sun is partially visible. There is a noticeable amount of smoke or haze rising from the right side of the image, which appears to be coming from industrial sources, as indicated by the presence of power lines and what looks like a factory chimney. The water reflects the sky's colors, creating a mirror-like effect. In the foreground, there is a grassy area with a few people walking along a path. The overall atmosphere is calm, with a mix of natural beauty and industrial activity."}
+{"image_path": "image/Va4djcAIDTBHxkEWh3sR9.jpeg", "image_description": "The image shows an outdoor scene with a person standing near a fence, pointing towards a large plume of dark smoke rising into the sky. The smoke appears to be coming from a distant area, possibly indicating a fire or some other industrial activity. The person is dressed casually and is standing on grassy terrain with a clear blue sky above. There are trees and open fields in the background, suggesting a rural or semi-rural setting. The smoke is thick and dark, contrasting sharply with the bright sky."}
+{"image_path": "image/Mlge558eDQglMeFC2d-tT.jpeg", "image_description": "The image shows a view from inside a vehicle on a road with multiple lanes. There are several cars visible ahead, and the road appears to be in an urban area with buildings on both sides. In the background, there are tall smokestacks emitting white smoke, which is indicative of industrial activity. The sky is clear, but the presence of the smoke suggests that air quality might be affected by the emissions from these stacks. The overall scene indicates a mix of urban life and industrial operations."}
+{"image_path": "image/ua_VWNMQkwRFILd4wgeIh.jpg", "image_description": "The image depicts an industrial area with a factory emitting smoke from its chimney. The sky is clear and blue, and there are mountains in the background, some of which appear to be partially covered with snow. The foreground shows a flat, open landscape with sparse vegetation.\nThe target \"smog\" in the image is the white plume of smoke rising from the factory's chimney. It is a significant feature in the scene, contrasting against the clear blue sky. The smoke appears dense and billowing, indicating that it is likely a result of industrial activity, possibly from burning fossil fuels or other materials. The presence of this smoke suggests potential air pollution, which can have negative impacts on the environment and human health."}
+{"image_path": "image/cX0RIMVZZuceLikcep7zU.jpeg", "image_description": "The image shows an urban street scene with a few people walking on the sidewalk and some vehicles parked or moving in the background. There is a noticeable plume of black smoke rising from a point near the center of the image, indicating a fire. The smoke is thick and dark, suggesting it could be from a vehicle or another source that has caught fire. The surrounding area appears to be a mix of greenery and buildings, with some construction cranes visible in the distance, indicating ongoing development. The weather seems overcast, contributing to the visibility of the smoke."}
+{"image_path": "image/DhvK2RDK6-G9WaLvhqYhe.jpg", "image_description": "The image shows a nighttime scene with a building that appears to be on fire or experiencing some form of smoke or fire activity. There is a significant amount of smoke rising from the upper floors of the building, indicating a potential fire or other emergency situation. The building is illuminated by artificial lights, and the smoke is dense and dark, suggesting it could be thick and possibly toxic. The surrounding area is not clearly visible due to the darkness and the focus on the building and the smoke."}
+{"image_path": "image/onneNwxD9igAEwCt1zwOV.jpeg", "image_description": "The image depicts an industrial area located near a body of water during what appears to be either dawn or dusk, as the sky has a soft blue hue and the lights from the industrial structures are illuminated. The industrial complex consists of numerous tall towers and cylindrical tanks, which are characteristic of a petrochemical or oil refinery facility. There are visible plumes of smoke rising from some of the structures, indicating ongoing industrial activity.\nThe water in the foreground is calm, reflecting the lights from the industrial buildings and creating a mirror-like effect on the surface. In the background, there are hills or low mountains, adding depth to the scene. The overall atmosphere is serene yet industrious, with the contrast between the natural landscape and the man-made structures being quite pronounced.\nRegarding the presence of smog, it can be observed in the image. The plumes of smoke rising from the industrial structures suggest that there is indeed smog present. The smoke appears to be thick and white, indicating the emission of pollutants such as sulfur dioxide, nitrogen oxides, and particulate matter. This type of pollution is typical of industrial facilities and can contribute to air quality issues in the surrounding areas."}
+{"image_path": "image/-NJ5xBiM8i-ThNAVKGbYG.jpeg", "image_description": "The image shows an industrial area with several tall smokestacks emitting white smoke into the sky. The sky appears hazy, suggesting the presence of smog. There are multiple vehicles on the road in the foreground, and some buildings are visible in the background. The overall atmosphere seems to be affected by air pollution, as indicated by the smog and the visible emissions from the smokestacks."}
+{"image_path": "image/4mBy6LRkPucdhu7XxgEot.jpg", "image_description": "The image depicts an industrial area with multiple smokestacks emitting large plumes of white smoke into the air. The scene appears to be taken during either sunrise or sunset, as indicated by the soft lighting and the gradient in the sky transitioning from light to darker hues. The industrial complex includes various buildings and structures, some of which have visible chimneys or stacks. The smoke is thick and billowing, suggesting significant industrial activity, possibly related to manufacturing or energy production. The surrounding area seems to be relatively flat, with no prominent natural features like trees or mountains in the immediate vicinity. The overall atmosphere suggests a focus on heavy industry and potential environmental impact due to the emissions."}
+{"image_path": "image/vWF4bdevHe3GqcGtz8Iwv.jpg", "image_description": "The image shows a body of water in the foreground with a cityscape in the background. There are several high-rise buildings visible, and smoke can be seen rising from one of the buildings on the right side of the image. The sky appears clear, and there is no significant smog present in the image. The smoke is localized to the area near the building and does not extend across the entire scene."}
+{"image_path": "image/t09pz0CWWNaOTkC1B64SK.jpg", "image_description": "The image depicts an industrial complex with multiple smokestacks emitting plumes of white smoke into the sky. The complex appears to be a power plant or chemical facility, characterized by large buildings and infrastructure. The surrounding area includes greenery, with trees and grass visible on both sides of the road that runs through the facility.\nRegarding the presence of smog, the image shows significant emissions from the smokestacks, which could contribute to air pollution. However, the term \"smog\" typically refers to a mixture of pollutants in the atmosphere, often including particulate matter, nitrogen oxides, and sulfur dioxide, which can lead to poor air quality. In this image, the visible emissions suggest potential air pollution, but without specific data on the composition of these emissions, it's not possible to definitively classify them as smog. The image does highlight the environmental impact of industrial activities on air quality."}
diff --git a/benchmark/smog/user_input.txt b/benchmark/smog/user_input.txt
new file mode 100644
index 0000000000000000000000000000000000000000..a11782f9adaece3538cf974526f149ea055e3616
--- /dev/null
+++ b/benchmark/smog/user_input.txt
@@ -0,0 +1 @@
+检测烟雾
\ No newline at end of file
diff --git "a/benchmark/smog/\345\217\202\350\200\203\345\233\276\347\211\207/76d3682c-74f3-4fa7-a75a-94a225c55231.jpg" "b/benchmark/smog/\345\217\202\350\200\203\345\233\276\347\211\207/76d3682c-74f3-4fa7-a75a-94a225c55231.jpg"
new file mode 100644
index 0000000000000000000000000000000000000000..49e89b4a753bee184f643c5551bebf78e6ce655e
Binary files /dev/null and "b/benchmark/smog/\345\217\202\350\200\203\345\233\276\347\211\207/76d3682c-74f3-4fa7-a75a-94a225c55231.jpg" differ
diff --git a/benchmark/water/gt.jsonl b/benchmark/water/gt.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..130a084b97cce7c2a3dd4a8e1a5932743900bf95
--- /dev/null
+++ b/benchmark/water/gt.jsonl
@@ -0,0 +1,30 @@
+{"filename": "image/IkKMCU1PTdFGeAclmetow.jpg", "image_height": 562, "image_width": 1000, "instances": [{"label": 1, "label_name": "water", "bbox": [56, 250, 809, 554]}]}
+{"filename": "image/D1hpa5bU7fqTH3w8Ihqlt.jpeg", "image_height": 300, "image_width": 450, "instances": [{"label": 1, "label_name": "water", "bbox": [89, 91, 447, 298]}]}
+{"filename": "image/9zQR3UTnc106GQMYliRIf.webp", "image_height": 721, "image_width": 960, "instances": [{"label": 1, "label_name": "water", "bbox": [6, 351, 952, 716]}]}
+{"filename": "image/QoPAAm9xBhaQ2bwGYE4s1.webp", "image_height": 720, "image_width": 960, "instances": [{"label": 1, "label_name": "water", "bbox": [8, 366, 950, 714]}]}
+{"filename": "image/CNy0Nl14862JRCa1gOeYZ.jpg", "image_height": 550, "image_width": 550, "instances": [{"label": 1, "label_name": "water", "bbox": [17, 464, 541, 546]}]}
+{"filename": "image/D3l_pT9SoB5NrTPJHey8e.jpg", "image_height": 400, "image_width": 600, "instances": [{"label": 1, "label_name": "water", "bbox": [6, 294, 309, 394]}]}
+{"filename": "image/CzNfn-KcVwwT94-0jKuvH.png", "image_height": 1051, "image_width": 1429, "instances": [{"label": 1, "label_name": "water", "bbox": [167, 538, 1406, 1024]}]}
+{"filename": "image/F9Jp_THhd1JDVHDwIfyvU.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [142, 322, 1894, 1024]}, {"label": 1, "label_name": "water", "bbox": [1426, 800, 1640, 900]}, {"label": 1, "label_name": "water", "bbox": [1614, 890, 1914, 995]}]}
+{"filename": "image/XO1mN5u0WvPP-38vQaq7o.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [294, 406, 1688, 868]}]}
+{"filename": "image/h4-H07qLuHKXH5UX-h3xZ.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [24, 230, 1341, 793]}]}
+{"filename": "image/OSDdTyVcBa6lMfIJhDOgS.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [18, 279, 1893, 999]}]}
+{"filename": "image/J5MeAlZD__Sq5QcJyf4gv.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [56, 189, 1453, 1058]}]}
+{"filename": "image/DF6-H_sErOB8BgG6VDcgj.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [596, 223, 1828, 998]}]}
+{"filename": "image/ikVoHKGpwXFG2LmWF1Usw.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [18, 193, 1903, 1054]}]}
+{"filename": "image/tER9XZiQ1Ley4S3UfpvmP.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [16, 80, 1816, 973]}]}
+{"filename": "image/6VsaAYiP1DGftITp0-CMc.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [20, 165, 1903, 1065]}]}
+{"filename": "image/AFPldc8N0Wt6VqquthNYc.jpg", "image_height": 1080, "image_width": 1920, "instances": [{"label": 1, "label_name": "water", "bbox": [453, 206, 1914, 1069]}]}
+{"filename": "image/r9kghzfn9pMJBVjvBUsSJ.PNG", "image_height": 870, "image_width": 1614, "instances": [{"label": 1, "label_name": "water", "bbox": [248, 227, 841, 826]}]}
+{"filename": "image/TUEL3vwuX7nxaI-P0RVZs.PNG", "image_height": 875, "image_width": 1607, "instances": [{"label": 1, "label_name": "water", "bbox": [752, 16, 1587, 768]}]}
+{"filename": "image/I9tzSGkcPQqvg9iKeJmxH.PNG", "image_height": 887, "image_width": 1610, "instances": [{"label": 1, "label_name": "water", "bbox": [564, 324, 852, 848]}]}
+{"filename": "image/5PfcirXJfTg3STKLMXt3Z.PNG", "image_height": 878, "image_width": 1622, "instances": [{"label": 1, "label_name": "water", "bbox": [401, 579, 859, 852]}, {"label": 1, "label_name": "water", "bbox": [906, 602, 1100, 842]}, {"label": 1, "label_name": "water", "bbox": [583, 345, 841, 579]}]}
+{"filename": "image/cYe1ZST7bD0l5K3hrSIsY.PNG", "image_height": 871, "image_width": 1604, "instances": [{"label": 1, "label_name": "water", "bbox": [98, 396, 569, 784]}, {"label": 1, "label_name": "water", "bbox": [253, 178, 374, 302]}, {"label": 1, "label_name": "water", "bbox": [109, 301, 287, 375]}]}
+{"filename": "image/54QUfL1zu9KxdgC0Lnkaq.PNG", "image_height": 874, "image_width": 1602, "instances": [{"label": 1, "label_name": "water", "bbox": [421, 663, 520, 857]}, {"label": 1, "label_name": "water", "bbox": [731, 566, 1044, 704]}]}
+{"filename": "image/IS_8jyoAarGlGjEQPIMBt.PNG", "image_height": 863, "image_width": 1616, "instances": [{"label": 1, "label_name": "water", "bbox": [1199, 499, 1571, 735]}]}
+{"filename": "image/-HriUP8AIMoxb6ARMUqe_.PNG", "image_height": 870, "image_width": 1614, "instances": [{"label": 1, "label_name": "water", "bbox": [509, 176, 694, 391]}, {"label": 1, "label_name": "water", "bbox": [1188, 371, 1421, 541]}]}
+{"filename": "image/9Tq31VYy_-FXffXrf8wrC.PNG", "image_height": 879, "image_width": 1619, "instances": [{"label": 1, "label_name": "water", "bbox": [977, 292, 1129, 549]}]}
+{"filename": "image/Ji5jAGAaWpGp7l5nPgfCa.PNG", "image_height": 879, "image_width": 1620, "instances": [{"label": 1, "label_name": "water", "bbox": [693, 234, 1367, 821]}, {"label": 1, "label_name": "water", "bbox": [1366, 521, 1583, 675]}]}
+{"filename": "image/Cya8krjEz3_toFmOyU4K4.webp", "image_height": 667, "image_width": 500, "instances": [{"label": 1, "label_name": "water", "bbox": [160, 368, 491, 654]}]}
+{"filename": "image/lF6gCeJ6sT9rPN5MMeJcz.webp", "image_height": 1067, "image_width": 800, "instances": [{"label": 1, "label_name": "water", "bbox": [17, 586, 491, 927]}]}
+{"filename": "image/NNIXbhHAZYkKp8sRtkUo3.jpg", "image_height": 750, "image_width": 1000, "instances": [{"label": 1, "label_name": "water", "bbox": [260, 263, 980, 741]}]}
diff --git a/benchmark/water/image/-HriUP8AIMoxb6ARMUqe_.PNG b/benchmark/water/image/-HriUP8AIMoxb6ARMUqe_.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..cc804086ac6f47adfddc155896903dd318670cf9
--- /dev/null
+++ b/benchmark/water/image/-HriUP8AIMoxb6ARMUqe_.PNG
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a576e38565cde403872828976c2ddbe62104f6f322655967b6bbaefb87fa84f
+size 2839798
diff --git a/benchmark/water/image/54QUfL1zu9KxdgC0Lnkaq.PNG b/benchmark/water/image/54QUfL1zu9KxdgC0Lnkaq.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..25fc1ba712b178c70645ec84d7032e89253c704b
--- /dev/null
+++ b/benchmark/water/image/54QUfL1zu9KxdgC0Lnkaq.PNG
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90d3cb0a94a7926439429238ad9b1dc16504a5ca0fc80faf2531600044f5b9d4
+size 2641621
diff --git a/benchmark/water/image/5PfcirXJfTg3STKLMXt3Z.PNG b/benchmark/water/image/5PfcirXJfTg3STKLMXt3Z.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..2d62db3ea8d60ac1fa43a90909be349b32f34fd3
--- /dev/null
+++ b/benchmark/water/image/5PfcirXJfTg3STKLMXt3Z.PNG
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ef058fef868109f302611ecfc18102ac1fdcfb6a44050fa0487ea751dbd3572
+size 2917494
diff --git a/benchmark/water/image/6VsaAYiP1DGftITp0-CMc.jpg b/benchmark/water/image/6VsaAYiP1DGftITp0-CMc.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..b329b230b077b9987e9a1f120e8ca25e8a6a9a8d
--- /dev/null
+++ b/benchmark/water/image/6VsaAYiP1DGftITp0-CMc.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7aebb2273eaec4c876584716e897beec14481c8aa2e8c34d1149388c2eae7c9
+size 190933
diff --git a/benchmark/water/image/9Tq31VYy_-FXffXrf8wrC.PNG b/benchmark/water/image/9Tq31VYy_-FXffXrf8wrC.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..04fccc5fea67150f2f0dcedeb6ea93e90fbfa477
--- /dev/null
+++ b/benchmark/water/image/9Tq31VYy_-FXffXrf8wrC.PNG
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bff8f300c56cef660bd47fd5503097f6f7c6ce9fc0c9bc253c984b73b38f8b8
+size 3576706
diff --git a/benchmark/water/image/9zQR3UTnc106GQMYliRIf.webp b/benchmark/water/image/9zQR3UTnc106GQMYliRIf.webp
new file mode 100644
index 0000000000000000000000000000000000000000..fb5a83f733483d5dd2e87574d04c47f1c6c4f2ab
--- /dev/null
+++ b/benchmark/water/image/9zQR3UTnc106GQMYliRIf.webp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6be3da078e66844289084ec5ada94a68ce1e700fd7077a3904069d22f9ef8c5b
+size 185678
diff --git a/benchmark/water/image/AFPldc8N0Wt6VqquthNYc.jpg b/benchmark/water/image/AFPldc8N0Wt6VqquthNYc.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..e3bc246bedbfbd60d1c37f8d00258ec7fac00c1f
--- /dev/null
+++ b/benchmark/water/image/AFPldc8N0Wt6VqquthNYc.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8e5097b98328c339ae951a23696c6717cc9b7f25833a77ccf1343fceec8c404
+size 222167
diff --git a/benchmark/water/image/CNy0Nl14862JRCa1gOeYZ.jpg b/benchmark/water/image/CNy0Nl14862JRCa1gOeYZ.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..16845a47e912cf6a2d3566e8c6f2300b9f48fcec
Binary files /dev/null and b/benchmark/water/image/CNy0Nl14862JRCa1gOeYZ.jpg differ
diff --git a/benchmark/water/image/Cya8krjEz3_toFmOyU4K4.webp b/benchmark/water/image/Cya8krjEz3_toFmOyU4K4.webp
new file mode 100644
index 0000000000000000000000000000000000000000..eecbb930df04ae4ab3e47263c856d86d4774e40f
Binary files /dev/null and b/benchmark/water/image/Cya8krjEz3_toFmOyU4K4.webp differ
diff --git a/benchmark/water/image/CzNfn-KcVwwT94-0jKuvH.png b/benchmark/water/image/CzNfn-KcVwwT94-0jKuvH.png
new file mode 100644
index 0000000000000000000000000000000000000000..fa00afeafd4c062f67e8da7ba046e0a302adb6e7
--- /dev/null
+++ b/benchmark/water/image/CzNfn-KcVwwT94-0jKuvH.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e19d8b348233181d22c5e210e3be8c463f9b18947c4cac35e6ad15553329adf
+size 1756578
diff --git a/benchmark/water/image/D1hpa5bU7fqTH3w8Ihqlt.jpeg b/benchmark/water/image/D1hpa5bU7fqTH3w8Ihqlt.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..a3abe4339d3ec7e44550e73e1bf9065df074e0c5
Binary files /dev/null and b/benchmark/water/image/D1hpa5bU7fqTH3w8Ihqlt.jpeg differ
diff --git a/benchmark/water/image/D3l_pT9SoB5NrTPJHey8e.jpg b/benchmark/water/image/D3l_pT9SoB5NrTPJHey8e.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..2bbc4aa0b68b87d990a0e8ae2b34f04c6e923af7
--- /dev/null
+++ b/benchmark/water/image/D3l_pT9SoB5NrTPJHey8e.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c6510cbdc2715e47f7ced76d4f8c7c5a141fc78c477720683aca5b830e421fc
+size 113950
diff --git a/benchmark/water/image/DF6-H_sErOB8BgG6VDcgj.jpg b/benchmark/water/image/DF6-H_sErOB8BgG6VDcgj.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..24bcd76b1506029648b22745171387767fcbaa71
--- /dev/null
+++ b/benchmark/water/image/DF6-H_sErOB8BgG6VDcgj.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bf37d0827e80ac7a0c2819ab553f3c1c31d6ea90c698ae1b32d8bf90396e2c9
+size 202501
diff --git a/benchmark/water/image/F9Jp_THhd1JDVHDwIfyvU.jpg b/benchmark/water/image/F9Jp_THhd1JDVHDwIfyvU.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8a31b8f9c99f02974f52f634f3a857ac91a5b325
--- /dev/null
+++ b/benchmark/water/image/F9Jp_THhd1JDVHDwIfyvU.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:986b80b06756b4ad9e33f560f5b21936c4e913b203fb0cf805a49332457186b7
+size 239153
diff --git a/benchmark/water/image/I9tzSGkcPQqvg9iKeJmxH.PNG b/benchmark/water/image/I9tzSGkcPQqvg9iKeJmxH.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..264e9f5f52f8690fba71779b98ccb9ce9548d560
--- /dev/null
+++ b/benchmark/water/image/I9tzSGkcPQqvg9iKeJmxH.PNG
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8addc4485befd551c71a28bb276a5eda0328f1a52347911310e6091e983d512a
+size 2984077
diff --git a/benchmark/water/image/IS_8jyoAarGlGjEQPIMBt.PNG b/benchmark/water/image/IS_8jyoAarGlGjEQPIMBt.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..c2dce87e95041a402fcc01a9719acc0dd5070810
--- /dev/null
+++ b/benchmark/water/image/IS_8jyoAarGlGjEQPIMBt.PNG
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fcb4c4eba5dc3b0af4bf11bcc217482d28205b9be7f7512a31a6dcb8686e57b
+size 2438144
diff --git a/benchmark/water/image/IkKMCU1PTdFGeAclmetow.jpg b/benchmark/water/image/IkKMCU1PTdFGeAclmetow.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..3e436a25645cc651461615cf2a2e3215b7c74643
--- /dev/null
+++ b/benchmark/water/image/IkKMCU1PTdFGeAclmetow.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b3d129feab27f9a71b1e79ae339282984c3cd4387eac4a9f59c9780d0f46ad9
+size 146732
diff --git a/benchmark/water/image/J5MeAlZD__Sq5QcJyf4gv.jpg b/benchmark/water/image/J5MeAlZD__Sq5QcJyf4gv.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..0e53b63c7a37ba85b00a5ea95f4346cadaefff1f
--- /dev/null
+++ b/benchmark/water/image/J5MeAlZD__Sq5QcJyf4gv.jpg
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8cbb77f14b315601baabbd504d81509a0eb4d07c71334d6cd80b7592a0b029f
+size 161693
diff --git a/benchmark/water/image/Ji5jAGAaWpGp7l5nPgfCa.PNG b/benchmark/water/image/Ji5jAGAaWpGp7l5nPgfCa.PNG
new file mode 100644
index 0000000000000000000000000000000000000000..b7422e69698814eb8ab3649a9316608bcb6e1cc5
--- /dev/null
+++ b/benchmark/water/image/Ji5jAGAaWpGp7l5nPgfCa.PNG
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:776eea496b5c4dae8577f56bd05964626bb68ed49fa46f7b3ed483fa97043941
+size 3404446