id
unknown
images
listlengths
3
3
⌀
conversations
listlengths
2
2
source_dataset
stringclasses
7 values
image
listlengths
1
8
⌀
relative_point
listlengths
1
812
⌀
"object_ref/1197022961-bf915d8_cam02_obj13-obj39_left"
null
[ { "from": "human", "value": "<image>\nThe image displays an item encircled by a red rectangle. Locate several points on an item situated to the left side of the encircled item. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give y...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1197022961-bf915d8_cam02_obj13-obj39_left/image_0.png" ]
[ "<tag_name name=\"item situated to the left side of the encircled item\" img_idx=\"0\" point=\"[540, 879]\">", "<tag_name name=\"encircled item\" img_idx=\"0\" point=\"[616, 879]\">" ]
"refcoco_2439"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: red. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> and </answer> t...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_2439/image_0.jpg" ]
[ "<tag_name name=\"red jacket\" img_idx=\"0\" point=\"[658, 391]\">", "<tag_name name=\"red goggles\" img_idx=\"0\" point=\"[658, 204]\">", "<tag_name name=\"red ski boots\" img_idx=\"0\" point=\"[784, 870]\">" ]
"object_ref/1123113071-fe2f065_cam06_obj9-obj10_behind"
null
[ { "from": "human", "value": "<image>\nThe image features an item encased within a red rectangular border. Locate at least a single point on an item situated behind the object that has been highlighted by the rectangle. The answer should be presented in JSON format. Provide your thinking process between the ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1123113071-fe2f065_cam06_obj9-obj10_behind/image_0.png" ]
[ "<tag_name name=\"white mug\" img_idx=\"0\" point=\"[385, 907]\">", "<tag_name name=\"blue ball\" img_idx=\"0\" point=\"[464, 917]\">", "<tag_name name=\"green cube\" img_idx=\"0\" point=\"[268, 851]\">", "<tag_name name=\"orange cabinet\" img_idx=\"0\" point=\"[347, 601]\">", "<tag_name name=\"white mug\" ...
"refcoco_3552"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: left bike. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> and </ans...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_3552/image_0.jpg" ]
[ "<tag_name name=\"left bike\" img_idx=\"0\" point=\"[278, 460]\">" ]
"roborefit_34723"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: will you please fetch me the conical bottled facial cleanser. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then gi...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_34723/image_0.png" ]
[ "<tag_name name=\"conical bottled facial cleanser\" img_idx=\"0\" point=\"[67, 385]\">" ]
"refcoco_10617"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: girl in tanktop leaning. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <ans...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_10617/image_0.jpg" ]
[ "<tag_name name=\"girl in tanktop leaning\" img_idx=\"0\" point=\"[67, 351]\">" ]
"refcoco_14328"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: person wearing 147. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> ...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_14328/image_0.jpg" ]
[ "<tag_name name=\"person wearing 147\" img_idx=\"0\" point=\"[671, 548]\">" ]
"object_ref/1298394308-fe2f065_cam03_obj15-wall_cabinet_1-corpus_inside"
null
[ { "from": "human", "value": "<image>\nThe image features a container delineated by a red rectangular border. Pinpoint several spots on an item located within the confines of the marked container. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> ta...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1298394308-fe2f065_cam03_obj15-wall_cabinet_1-corpus_inside/image_0.png" ]
[ "<tag_name name=\"cabinet\" img_idx=\"0\" point=\"[650, 278]\">" ]
"region_ref/1291913914-8db35a4_cam04_obj15-obj12_beside"
null
[ { "from": "human", "value": "<image>\nThe image features an item encircled by a red rectangle. Determine one or more locations within a vacant area next to the encircled item. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give yo...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1291913914-8db35a4_cam04_obj15-obj12_beside/image_0.png" ]
[ "<tag_name name=\"encircled item\" img_idx=\"0\" point=\"[745, 609]\">" ]
"fsd_pointbridge_105306"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: put the fork to the orange pot. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer should be prese...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_105306/image_0.jpg" ]
[ "<tag_name name=\"fork\" img_idx=\"0\" point=\"[341, 658]\">", "<tag_name name=\"orange pot\" img_idx=\"0\" point=\"[708, 504]\">" ]
"object_ref/1247447187-bf915d8_cam04_obj33-obj19_front"
null
[ { "from": "human", "value": "<image>\nThe image features an item encircled by a red rectangle. Locate several spots on an item that is situated ahead of the encircled item. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1247447187-bf915d8_cam04_obj33-obj19_front/image_0.png" ]
[ "<tag_name name=\"encircled item\" img_idx=\"0\" point=\"[846, 319]\">", "<tag_name name=\"item situated ahead of the encircled item\" img_idx=\"0\" point=\"[874, 357]\">" ]
"roborefit_12487"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: hand me the purple cylindrical bottled vitamins. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_12487/image_0.png" ]
[ "<tag_name name=\"purple cylindrical bottled vitamins\" img_idx=\"0\" point=\"[718, 326]\">" ]
"region_ref/1311983478-1ed231c_cam10_obj30-sink_cabinet-sink_countertop_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area delineated by a red rectangle. Locate a few points within the vacant area that is situated on the outlined surface. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then gi...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1311983478-1ed231c_cam10_obj30-sink_cabinet-sink_countertop_on/image_0.png" ]
[ "<tag_name name=\"white countertop, red rectangle outline\" img_idx=\"0\" point=\"[874, 403]\">" ]
"refcoco_5384"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: kid with hand up near his hair. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between t...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_5384/image_0.jpg" ]
[ "<tag_name name=\"kid with hand up near his hair\" img_idx=\"0\" point=\"[240, 317]\">" ]
"region_ref/121120346-f2789f5_cam06_obj33-sink_cabinet-sink_countertop_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area demarcated by a red rectangle. Locate several points within a vacant space on the marked area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/121120346-f2789f5_cam06_obj33-sink_cabinet-sink_countertop_on/image_0.png" ]
[ "<tag_name name=\"sink\" img_idx=\"0\" point=\"[405, 361]\">", "<tag_name name=\"wooden tray\" img_idx=\"0\" point=\"[539, 408]\">", "<tag_name name=\"plant\" img_idx=\"0\" point=\"[398, 317]\">" ]
"object_ref/1302791167-b55b742_cam04_obj39-obj36_beside"
null
[ { "from": "human", "value": "<image>\nThe image features an item encircled by a red rectangular border. Determine one or more specific locations on an item that is adjacent to the one encased by the red rectangle. The answer should be presented in JSON format. Provide your thinking process between the <thin...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1302791167-b55b742_cam04_obj39-obj36_beside/image_0.png" ]
[ "<tag_name name=\"white rectangular box\" img_idx=\"0\" point=\"[507, 517]\">", "<tag_name name=\"brown headphones\" img_idx=\"0\" point=\"[746, 538]\">" ]
"region_ref/1211130419-858c02c_cam02_obj30-base_cabinet-drawer_0_0_board_1_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts an area outlined by a red rectangle. Locate several points within the unoccupied space on the marked area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1211130419-858c02c_cam02_obj30-base_cabinet-drawer_0_0_board_1_on/image_0.png" ]
[ "<tag_name name=\"green shelf\" img_idx=\"0\" point=\"[513, 846]\">" ]
"object_ref/1101394943-cc5783a_cam07_obj33-oven-door_0_1_door_on"
null
[ { "from": "human", "value": "<image>\nThe image features a plane outlined by a red rectangular border. Locate one or more spots on an item that is situated upon the marked plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1101394943-cc5783a_cam07_obj33-oven-door_0_1_door_on/image_0.png" ]
[ "<tag_name name=\"white bowl\" img_idx=\"0\" point=\"[573, 714]\">", "<tag_name name=\"yellow paper\" img_idx=\"0\" point=\"[468, 760]\">", "<tag_name name=\"pink object with black and white pattern\" img_idx=\"0\" point=\"[458, 673]\">" ]
"fsd_pointbridge_87393"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: put knife in pot cardboard fence. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer should be pre...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_87393/image_0.jpg" ]
[ "<tag_name name=\"knife\" img_idx=\"0\" point=\"[398, 214]\">", "<tag_name name=\"pot\" img_idx=\"0\" point=\"[675, 470]\">", "<tag_name name=\"cardboard fence\" img_idx=\"0\" point=\"[501, 361]\">" ]
"roborefit_14567"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: would you please fetch me the polar bear. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_14567/image_0.png" ]
[ "<tag_name name=\"polar bear\" img_idx=\"0\" point=\"[738, 240]\">" ]
"fsd_pointrtx_5823"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: move orange can near orange. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer should be presente...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointrtx_5823/image_0.jpg" ]
[ "<tag_name name=\"orange can\" img_idx=\"0\" point=\"[640, 398]\">", "<tag_name name=\"orange\" img_idx=\"0\" point=\"[257, 234]\">" ]
"region_ref/131527571-4fb8b7b_cam10_obj10-oven-top_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts an area outlined by a red rectangle. Locate any number of points situated in a vacant space on the outlined area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your fin...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/131527571-4fb8b7b_cam10_obj10-oven-top_on/image_0.png" ]
[ "<tag_name name=\"vacant space\" img_idx=\"0\" point=\"[408, 632]\">" ]
"roborefit_4811"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: I need to get the conical bottled shampoo on the middle shelf. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then g...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_4811/image_0.png" ]
[ "<tag_name name=\"conical bottled shampoo\" img_idx=\"0\" point=\"[718, 564]\">" ]
"object_ref/1238442129-f53019d_cam06_obj6-refrigerator-shelf_2_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts a plane delineated by a red rectangular border. Locate several points situated on an item that rests upon the outlined plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then g...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1238442129-f53019d_cam06_obj6-refrigerator-shelf_2_on/image_0.png" ]
[ "<tag_name name=\"white box\" img_idx=\"0\" point=\"[587, 494]\">", "<tag_name name=\"black box\" img_idx=\"0\" point=\"[501, 496]\">" ]
"region_ref/1106465562-bf915d8_cam01_obj5-wall_cabinet_1-corpus_inside"
null
[ { "from": "human", "value": "<image>\nThe image features a vessel delineated by a red rectangle. Locate several spots within the vacant area enclosed by the outlined vessel. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1106465562-bf915d8_cam01_obj5-wall_cabinet_1-corpus_inside/image_0.png" ]
[ "<tag_name name=\"vessel\" img_idx=\"0\" point=\"[851, 194]\">", "<tag_name name=\"vessel\" img_idx=\"0\" point=\"[840, 196]\">" ]
"fsd_pointbridge_78004"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: put the green cloth to the center of the table. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_78004/image_0.jpg" ]
[ "<tag_name name=\"green cloth\" img_idx=\"0\" point=\"[258, 476]\">", "<tag_name name=\"table\" img_idx=\"0\" point=\"[501, 439]\">" ]
"object_ref/1150618300-efbbcd9_cam08_obj0-refrigerator-shelf_1_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts a plane delineated by a red rectangular border. Locate several points situated on an item that rests upon the outlined plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then g...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1150618300-efbbcd9_cam08_obj0-refrigerator-shelf_1_on/image_0.png" ]
[ "<tag_name name=\"remote control\" img_idx=\"0\" point=\"[438, 806]\">", "<tag_name name=\"white object (possibly a small box or device)\" img_idx=\"0\" point=\"[517, 831]\">", "<tag_name name=\"black object (possibly a case or electronic device)\" img_idx=\"0\" point=\"[406, 819]\">", "<tag_name name=\"remot...
"roborefit_17905"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: will you please give me the purple cylindrical bottled vitamins on the right. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> t...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_17905/image_0.png" ]
[ "<tag_name name=\"purple cylindrical bottled vitamins\" img_idx=\"0\" point=\"[618, 703]\">", "<tag_name name=\"table\" img_idx=\"0\" point=\"[544, 390]\">" ]
"fsd_pointbridge_78272"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: moved the orange cloth to the right burner. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer sho...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_78272/image_0.jpg" ]
[ "<tag_name name=\"orange cloth\" img_idx=\"0\" point=\"[380, 576]\">", "<tag_name name=\"right burner\" img_idx=\"0\" point=\"[548, 494]\">" ]
"refcoco_915"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: bleacher on either side of ears - the thin bleacher. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your f...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_915/image_0.jpg" ]
[ "<tag_name name=\"bleacher\" img_idx=\"0\" point=\"[148, 306]\">", "<tag_name name=\"bleacher\" img_idx=\"0\" point=\"[705, 306]\">" ]
"refcoco_8973"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: the couch in back. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> a...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_8973/image_0.jpg" ]
[ "<tag_name name=\"couch\" img_idx=\"0\" point=\"[750, 489]\">" ]
"object_ref/1231941339-bf915d8_cam05_obj38-oven-door_0_1_door_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area outlined by a red rectangle. Locate several points on an item that is placed on top of the marked area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your fina...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1231941339-bf915d8_cam05_obj38-oven-door_0_1_door_on/image_0.png" ]
[ "<tag_name name=\"purple cube\" img_idx=\"0\" point=\"[739, 450]\">", "<tag_name name=\"red tray\" img_idx=\"0\" point=\"[658, 489]\">" ]
"roborefit_32327"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: pass me the facial cleanser with a black handy pump. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your f...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_32327/image_0.png" ]
[ "<tag_name name=\"facial cleanser with a black handy pump\" img_idx=\"0\" point=\"[548, 371]\">" ]
"roborefit_13707"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: will you please hand me the avocado. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer betw...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_13707/image_0.png" ]
[ "<tag_name name=\"avocado\" img_idx=\"0\" point=\"[480, 593]\">" ]
"refcoco_14537"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: A young woman wearing a pink shirt and a white veil.. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your ...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_14537/image_0.jpg" ]
[ "<tag_name name=\"young woman\" img_idx=\"0\" point=\"[483, 750]\">", "<tag_name name=\"pink shirt\" img_idx=\"0\" point=\"[496, 750]\">", "<tag_name name=\"white veil\" img_idx=\"0\" point=\"[511, 609]\">" ]
"region_ref/1093375184-63bd2a7_cam01_obj8-obj2-obj1_between"
null
[ { "from": "human", "value": "<image>\nThe image displays an item encircled by a red rectangle, as well as another item encircled by a green rectangle. Locate and identify several spots within the vacant space that separates the objects surrounded by the rectangles. The answer should be presented in JSON for...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1093375184-63bd2a7_cam01_obj8-obj2-obj1_between/image_0.png" ]
[ "<tag_name name=\"white figurine\" img_idx=\"0\" point=\"[410, 268]\">", "<tag_name name=\"blue bowl\" img_idx=\"0\" point=\"[507, 433]\">" ]
"refcoco_12458"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: sorry. man behind woman. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <ans...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_12458/image_0.jpg" ]
[ "<tag_name name=\"man behind woman\" img_idx=\"0\" point=\"[298, 175]\">" ]
"object_ref/1083074146-d5d1346_cam04_obj13-oven-heater_0_0_on"
null
[ { "from": "human", "value": "<image>\nThe image features a plane outlined by a red rectangular border. Locate one or more spots on an item that is situated upon the marked plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1083074146-d5d1346_cam04_obj13-oven-heater_0_0_on/image_0.png" ]
[ "<tag_name name=\"wooden bowl\" img_idx=\"0\" point=\"[71, 408]\">", "<tag_name name=\"green bottle\" img_idx=\"0\" point=\"[153, 306]\">", "<tag_name name=\"white paper object\" img_idx=\"0\" point=\"[349, 356]\">", "<tag_name name=\"small brown block\" img_idx=\"0\" point=\"[178, 548]\">", "<tag_name name...
"roborefit_7382"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: could you please give me the yellow bottled yogurt on the bottom left shelf. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> ta...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_7382/image_0.png" ]
[ "<tag_name name=\"yellow bottled yogurt\" img_idx=\"0\" point=\"[324, 816]\">" ]
"fsd_pointbridge_26232"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: move red pot to the right side of the sink. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer sho...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_26232/image_0.jpg" ]
[ "<tag_name name=\"red pot\" img_idx=\"0\" point=\"[650, 439]\">", "<tag_name name=\"sink\" img_idx=\"0\" point=\"[871, 548]\">" ]
"region_ref/1119222684-b55b742_cam05_obj6-obj0_right"
null
[ { "from": "human", "value": "<image>\nWithin the image, there is an item encircled by a red rectangle. Determine and select several locations in the area that is devoid of content and situated to the immediate right of the encircled item. The answer should be presented in JSON format. Provide your thinking ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1119222684-b55b742_cam05_obj6-obj0_right/image_0.png" ]
[ "<tag_name name=\"green square object with four black dots\" img_idx=\"0\" point=\"[287, 765]\">" ]
"refcoco_533"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: mid case. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> and </answ...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_533/image_0.jpg" ]
[ "<tag_name name=\"mid case\" img_idx=\"0\" point=\"[500, 648]\">" ]
"roborefit_31378"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: would you bring me the shampoo in a cylindrical bottle. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give you...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_31378/image_0.png" ]
[ "<tag_name name=\"shampoo in a cylindrical bottle\" img_idx=\"0\" point=\"[150, 749]\">" ]
"region_ref/1188883718-1cd1792_cam09_obj17-obj1_behind"
null
[ { "from": "human", "value": "<image>\nThe image features an item encircled by a red rectangle. Locate a few spots within a vacant area situated beyond the encased item. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your fina...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1188883718-1cd1792_cam09_obj17-obj1_behind/image_0.png" ]
[ "<tag_name name=\"encased item\" img_idx=\"0\" point=\"[347, 680]\">", "<tag_name name=\"encased item\" img_idx=\"0\" point=\"[341, 675]\">" ]
"object_ref/1099245379-8d3c67c_cam09_obj2-oven-heater_1_0_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts an area outlined by a red rectangle. Locate several points on any item that is situated on the outlined area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final a...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1099245379-8d3c67c_cam09_obj2-oven-heater_1_0_on/image_0.png" ]
[ "<tag_name name=\"red book\" img_idx=\"0\" point=\"[375, 107]\">", "<tag_name name=\"brown leather bag\" img_idx=\"0\" point=\"[410, 66]\">", "<tag_name name=\"white mug\" img_idx=\"0\" point=\"[398, 48]\">", "<tag_name name=\"purple pen\" img_idx=\"0\" point=\"[257, 109]\">", "<tag_name name=\"small round ...
"object_ref/1238827303-d5b363c_cam06_obj25-dishwasher-top_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts an area enclosed by a red rectangle. Locate several points on an item that is situated on the outlined area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final an...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1238827303-d5b363c_cam06_obj25-dishwasher-top_on/image_0.png" ]
[ "<tag_name name=\"brown box\" img_idx=\"0\" point=\"[531, 127]\">", "<tag_name name=\"white surface\" img_idx=\"0\" point=\"[486, 150]\">" ]
"object_ref/1315430684-8d3c67c_cam02_obj35-dishwasher-surface_0_1_on"
null
[ { "from": "human", "value": "<image>\nThe image features a plane outlined by a red rectangle. Locate several points on any item that is situated on the marked plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final a...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1315430684-8d3c67c_cam02_obj35-dishwasher-surface_0_1_on/image_0.png" ]
[ "<tag_name name=\"wooden cabinet\" img_idx=\"0\" point=\"[785, 341]\">", "<tag_name name=\"yellow object\" img_idx=\"0\" point=\"[769, 300]\">" ]
"roborefit_10569"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: can you find the symmetrical scissor. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer bet...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_10569/image_0.png" ]
[ "<tag_name name=\"scissors\" img_idx=\"0\" point=\"[587, 260]\">", "<tag_name name=\"scissors\" img_idx=\"0\" point=\"[714, 530]\">" ]
"fsd_pointbridge_42999"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: put the red figure on the top edge of the cloth. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answe...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_42999/image_0.jpg" ]
[ "<tag_name name=\"red figure\" img_idx=\"0\" point=\"[274, 450]\">", "<tag_name name=\"cloth\" img_idx=\"0\" point=\"[568, 531]\">" ]
"fsd_pointbridge_68027"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: put the blue arch on top of the green arch. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer sho...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_68027/image_0.jpg" ]
[ "<tag_name name=\"blue arch\" img_idx=\"0\" point=\"[301, 234]\">", "<tag_name name=\"green arch\" img_idx=\"0\" point=\"[406, 417]\">" ]
"fsd_pointbridge_11414"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: Move the cheese to the left of the pan.. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer should...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_11414/image_0.jpg" ]
[ "<tag_name name=\"cheese\" img_idx=\"0\" point=\"[630, 471]\">", "<tag_name name=\"pan\" img_idx=\"0\" point=\"[408, 375]\">" ]
"refcoco_6735"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: green. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> and </answer>...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_6735/image_0.jpg" ]
[ "<tag_name name=\"green t-shirt\" img_idx=\"0\" point=\"[385, 401]\">" ]
"refcoco_11940"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: dog at 9:00. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> and </a...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_11940/image_0.jpg" ]
[ "<tag_name name=\"dog at 9:00\" img_idx=\"0\" point=\"[371, 580]\">", "<tag_name name=\"hot dog bun\" img_idx=\"0\" point=\"[246, 419]\">", "<tag_name name=\"sausage\" img_idx=\"0\" point=\"[371, 580]\">" ]
"object_ref/1115074743-1ed231c_cam06_obj31-refrigerator-top_on"
null
[ { "from": "human", "value": "<image>\nThe image features a plane outlined by a red rectangular border. Locate one or more spots on an item that is situated upon the marked plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1115074743-1ed231c_cam06_obj31-refrigerator-top_on/image_0.png" ]
[ "<tag_name name=\"red rectangular border\" img_idx=\"0\" point=\"[894, 194]\">", "<tag_name name=\"wooden cabinet\" img_idx=\"0\" point=\"[750, 533]\">", "<tag_name name=\"white drawers\" img_idx=\"0\" point=\"[940, 466]\">" ]
"object_ref/1311840255-ab687ea_cam04_obj20-dishwasher-top_on"
null
[ { "from": "human", "value": "<image>\nThe image features a plane outlined by a red rectangular border. Locate any number of spots situated on an item that rests on the marked plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then g...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1311840255-ab687ea_cam04_obj20-dishwasher-top_on/image_0.png" ]
[ "<tag_name name=\"plate\" img_idx=\"0\" point=\"[587, 466]\">", "<tag_name name=\"chopsticks\" img_idx=\"0\" point=\"[450, 459]\">" ]
"object_ref/1204937237-d5d1346_cam02_obj4-sink_cabinet-sink_countertop_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area outlined by a red rectangle. Locate several points on an item that is placed on top of the marked area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your fina...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1204937237-d5d1346_cam02_obj4-sink_cabinet-sink_countertop_on/image_0.png" ]
[ "<tag_name name=\"yellow rectangular object\" img_idx=\"0\" point=\"[894, 635]\">", "<tag_name name=\"pink circular-object\" img_idx=\"0\" point=\"[975, 673]\">" ]
"roborefit_873"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: grab the glue stick on the right. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_873/image_0.png" ]
[ "<tag_name name=\"glue stick on the right\" img_idx=\"0\" point=\"[675, 264]\">" ]
"roborefit_25782"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: I need to grasp the canned coffee. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer betwee...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_25782/image_0.png" ]
[ "<tag_name name=\"canned coffee\" img_idx=\"0\" point=\"[357, 319]\">" ]
"roborefit_23264"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: would you please give me the folding umbrella. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final a...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_23264/image_0.png" ]
[ "<tag_name name=\"folding umbrella\" img_idx=\"0\" point=\"[730, 548]\">" ]
"object_ref/1186913270-0d5e33c_cam02_obj31-obj2_left"
null
[ { "from": "human", "value": "<image>\nThe image features an item encircled by a red rectangle. Locate and point out several spots on an item that is situated to the left side of the item encased by the red rectangle. The answer should be presented in JSON format. Provide your thinking process between the <t...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1186913270-0d5e33c_cam02_obj31-obj2_left/image_0.png" ]
[ "<tag_name name=\"cat\" img_idx=\"0\" point=\"[347, 516]\">", "<tag_name name=\"purple rectangular object\" img_idx=\"0\" point=\"[380, 562]\">", "<tag_name name=\"gray box-like object\" img_idx=\"0\" point=\"[470, 516]\">" ]
"refcoco_3586"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: giraffe on right. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer> an...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_3586/image_0.jpg" ]
[ "<tag_name name=\"giraffe on right\" img_idx=\"0\" point=\"[805, 679]\">" ]
"roborefit_9897"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: I want to grasp the white electronic thermometer. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your fina...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_9897/image_0.png" ]
[ "<tag_name name=\"white electronic thermometer\" img_idx=\"0\" point=\"[750, 341]\">" ]
"roborefit_26678"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: I want to get the wired mouse. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between th...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_26678/image_0.png" ]
[ "<tag_name name=\"wired mouse\" img_idx=\"0\" point=\"[450, 267]\">" ]
"roborefit_20991"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: could you please hand me the screwdriver. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_20991/image_0.png" ]
[ "<tag_name name=\"screwdriver\" img_idx=\"0\" point=\"[750, 385]\">" ]
"fsd_pointbridge_11282"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: Put the red ball inside the bowl.. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer should be pr...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_11282/image_0.jpg" ]
[ "<tag_name name=\"red ball\" img_idx=\"0\" point=\"[450, 336]\">", "<tag_name name=\"bowl\" img_idx=\"0\" point=\"[341, 478]\">" ]
"object_ref/1086299592-8db35a4_cam00_obj23-obj14_beside"
null
[ { "from": "human", "value": "<image>\nThe image features an item that is enclosed within a red rectangular border. Locate a few points on an item that is beside the item with the red outline. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1086299592-8db35a4_cam00_obj23-obj14_beside/image_0.png" ]
[ "<tag_name name=\"black object with red outline\" img_idx=\"0\" point=\"[153, 479]\">", "<tag_name name=\"kitchen counter\" img_idx=\"0\" point=\"[107, 500]\">" ]
"region_ref/1216755242-0d5e33c_cam09_obj32-obj16_right"
null
[ { "from": "human", "value": "<image>\nThe image features an item encased within a red rectangle. Locate several points within the vacant area that is situated to the right of the item marked by the rectangle. The answer should be presented in JSON format. Provide your thinking process between the <think> an...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1216755242-0d5e33c_cam09_obj32-obj16_right/image_0.png" ]
[ "<tag_name name=\"item marked by the rectangle\" img_idx=\"0\" point=\"[801, 479]\">" ]
"region_ref/1335019510-63bd2a7_cam03_obj19-wall_cabinet_1-surface_0_0_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area demarcated by a red rectangle. Locate several points within a vacant space on the marked area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1335019510-63bd2a7_cam03_obj19-wall_cabinet_1-surface_0_0_on/image_0.png" ]
[ "<tag_name name=\"red rectangle\" img_idx=\"0\" point=\"[894, 206]\">" ]
"fsd_pointbridge_27194"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: put pot or pan from sink into drying rack. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer shou...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_27194/image_0.jpg" ]
[ "<tag_name name=\"pot or pan\" img_idx=\"0\" point=\"[573, 806]\">", "<tag_name name=\"drying rack\" img_idx=\"0\" point=\"[419, 306]\">" ]
"object_ref/120537581-b55b742_cam08_obj6-obj20_front"
null
[ { "from": "human", "value": "<image>\nThe image features an item encased within a red rectangular border. Locate at least a single point on an item that is situated in front of the item surrounded by the red outline. The answer should be presented in JSON format. Provide your thinking process between the <t...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/120537581-b55b742_cam08_obj6-obj20_front/image_0.png" ]
[ "<tag_name name=\"gray rectangular panel\" img_idx=\"0\" point=\"[375, 341]\">" ]
"region_ref/1236542068-cc5783a_cam03_obj8-obj14_beside"
null
[ { "from": "human", "value": "<image>\nIn the image, an item is enclosed within a red rectangular border. Locate several spots in the vacant area next to the marked item. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your fin...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1236542068-cc5783a_cam03_obj8-obj14_beside/image_0.png" ]
[ "<tag_name name=\"marked item\" img_idx=\"0\" point=\"[710, 507]\">" ]
"object_ref/1158445308-f2789f5_cam06_obj0-oven-heater_1_0_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area delineated by a red rectangle. Pinpoint one or several locations upon an item that is situated upon the marked area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then g...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1158445308-f2789f5_cam06_obj0-oven-heater_1_0_on/image_0.png" ]
[ "<tag_name name=\"red rectangle\" img_idx=\"0\" point=\"[216, 257]\">" ]
"fsd_pointbridge_62985"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: pick up pot. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer should be presented in JSON format...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointbridge_62985/image_0.jpg" ]
[ "<tag_name name=\"pot\" img_idx=\"0\" point=\"[587, 601]\">" ]
"object_ref/1057244821-cc5783a_cam03_obj37-base_cabinet-drawer_0_0_inside"
null
[ { "from": "human", "value": "<image>\nThe image features a vessel delineated by a red rectangular border. Locate a few points on any item situated within the confines of the marked vessel. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1057244821-cc5783a_cam03_obj37-base_cabinet-drawer_0_0_inside/image_0.png" ]
[ "<tag_name name=\"camera\" img_idx=\"0\" point=\"[375, 874]\">", "<tag_name name=\"stool\" img_idx=\"0\" point=\"[501, 769]\">", "<tag_name name=\"grid structure\" img_idx=\"0\" point=\"[546, 908]\">" ]
"region_ref/1058555033-d5b363c_cam05_obj9-dishwasher-door_0_1_door_on"
null
[ { "from": "human", "value": "<image>\nThe image features a plane marked by a red rectangular border. Locate a few points within an unoccupied area situated on the demarcated plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then gi...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1058555033-d5b363c_cam05_obj9-dishwasher-door_0_1_door_on/image_0.png" ]
[ "<tag_name name=\"red rectangular border\" img_idx=\"0\" point=\"[904, 935]\">" ]
"roborefit_22449"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: pass me the black and gray bluetooth mouse. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answ...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_22449/image_0.png" ]
[ "<tag_name name=\"black and gray bluetooth mouse\" img_idx=\"0\" point=\"[708, 714]\">" ]
"object_ref/1211084699-0d5e33c_cam05_obj23-refrigerator-freezer_separator_on"
null
[ { "from": "human", "value": "<image>\nThe image features a plane outlined by a red rectangular border. Locate one or more spots on an item that is situated upon the marked plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1211084699-0d5e33c_cam05_obj23-refrigerator-freezer_separator_on/image_0.png" ]
[ "<tag_name name=\"shelf\" img_idx=\"0\" point=\"[803, 364]\">" ]
"roborefit_20844"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: I would like to get the toy tiger. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer betwee...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_20844/image_0.png" ]
[ "<tag_name name=\"toy tiger\" img_idx=\"0\" point=\"[604, 658]\">" ]
"object_ref/1264591332-cc5783a_cam09_obj9-obj14_right"
null
[ { "from": "human", "value": "<image>\nThe image features an object encircled by a red rectangle. Determine several points on an object that is situated to the right of the encircled object. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, an...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1264591332-cc5783a_cam09_obj9-obj14_right/image_0.png" ]
[ "<tag_name name=\"box\" img_idx=\"0\" point=\"[513, 466]\">" ]
"refcoco_15272"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: Police Vehicle in rounding. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_15272/image_0.jpg" ]
[ "<tag_name name=\"police officer\" img_idx=\"0\" point=\"[540, 317]\">", "<tag_name name=\"police officer\" img_idx=\"0\" point=\"[168, 290]\">", "<tag_name name=\"police motorcycle\" img_idx=\"0\" point=\"[540, 638]\">", "<tag_name name=\"police motorcycle\" img_idx=\"0\" point=\"[147, 538]\">" ]
"refcoco_6011"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: plane behind one closest to camera. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer betwe...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_6011/image_0.jpg" ]
[ "<tag_name name=\"plane behind one closest to camera\" img_idx=\"0\" point=\"[351, 408]\">" ]
"object_ref/1138828097-fe2f065_cam06_obj35-dishwasher-top_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts an area enclosed by a red rectangle. Locate several points on an item that is situated on the outlined area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final an...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1138828097-fe2f065_cam06_obj35-dishwasher-top_on/image_0.png" ]
[ "<tag_name name=\"red rectangle\" img_idx=\"0\" point=\"[503, 976]\">" ]
"region_ref/1214988098-858c02c_cam05_obj10-sink_cabinet-corpus_inside"
null
[ { "from": "human", "value": "<image>\nThe image features a vessel delineated by a red rectangle. Locate several spots within the vacant area enclosed by the outlined vessel. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1214988098-858c02c_cam05_obj10-sink_cabinet-corpus_inside/image_0.png" ]
[ "<tag_name name=\"vessel\" img_idx=\"0\" point=\"[640, 375]\">" ]
"object_ref/1246326770-1cd1792_cam07_obj22-oven-heater_0_0_on"
null
[ { "from": "human", "value": "<image>\nThe image depicts a plane delineated by a red rectangular border. Locate several points situated on an item that rests upon the outlined plane. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then g...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1246326770-1cd1792_cam07_obj22-oven-heater_0_0_on/image_0.png" ]
[ "<tag_name name=\"sandwich maker\" img_idx=\"0\" point=\"[450, 76]\">" ]
"object_ref/1214445972-efbbcd9_cam08_obj8-oven-heater_1_0_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area outlined by a red rectangle. Locate one or several spots on an item that is positioned on the marked area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your f...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1214445972-efbbcd9_cam08_obj8-oven-heater_1_0_on/image_0.png" ]
[ "<tag_name name=\"red rectangle\" img_idx=\"0\" point=\"[507, 548]\">", "<tag_name name=\"circular object\" img_idx=\"0\" point=\"[513, 466]\">", "<tag_name name=\"kitchen counter\" img_idx=\"0\" point=\"[498, 593]\">", "<tag_name name=\"red rectangle\" img_idx=\"0\" point=\"[507, 548]\">", "<tag_name name=...
"object_ref/1039225686-858c02c_cam05_obj30-obj47_beside"
null
[ { "from": "human", "value": "<image>\nThe image features an item encircled by a red rectangular border. Determine one or more specific locations on an item that is adjacent to the one encased by the red rectangle. The answer should be presented in JSON format. Provide your thinking process between the <thin...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1039225686-858c02c_cam05_obj30-obj47_beside/image_0.png" ]
[ "<tag_name name=\"white chair\" img_idx=\"0\" point=\"[410, 479]\">", "<tag_name name=\"black circular object\" img_idx=\"0\" point=\"[385, 367]\">", "<tag_name name=\"wooden table\" img_idx=\"0\" point=\"[410, 658]\">" ]
"object_ref/1174307461-b55b742_cam00_obj8-obj18_right"
null
[ { "from": "human", "value": "<image>\nThe image depicts an item encircled by a red rectangle. Locate at least one point on an item situated to the right of the encircled item. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give yo...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1174307461-b55b742_cam00_obj8-obj18_right/image_0.png" ]
[ "<tag_name name=\"red rectangular object\" img_idx=\"0\" point=\"[503, 864]\">" ]
"refcoco_3756"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: guy left of umbrella. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answer...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_3756/image_0.jpg" ]
[ "<tag_name name=\"guy left of umbrella\" img_idx=\"0\" point=\"[507, 648]\">" ]
"refcoco_17240"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: The larger zebra who is on the grass.. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer be...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_17240/image_0.jpg" ]
[ "<tag_name name=\"zebra\" img_idx=\"0\" point=\"[745, 468]\">" ]
"object_ref/1045100730-ab687ea_cam04_obj35-base_cabinet-drawer_0_0_inside"
null
[ { "from": "human", "value": "<image>\nThe image features a container encircled by a red rectangle. Locate a few points on an item that resides within the boundaries of the marked container. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, an...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1045100730-ab687ea_cam04_obj35-base_cabinet-drawer_0_0_inside/image_0.png" ]
[ "<tag_name name=\"drawer\" img_idx=\"0\" point=\"[385, 744]\">", "<tag_name name=\"wooden chair\" img_idx=\"0\" point=\"[319, 780]\">", "<tag_name name=\"box\" img_idx=\"0\" point=\"[406, 736]\">" ]
"roborefit_34006"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: pass me the baking soda toothpaste in a big cyan conical bottle. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_34006/image_0.png" ]
[ "<tag_name name=\"baking soda toothpaste in a big cyan conical bottle\" img_idx=\"0\" point=\"[740, 759]\">" ]
"fsd_pointrtx_19318"
null
[ { "from": "human", "value": "<image>\nYou are currently a robot performing robotic manipulation tasks. The task instruction is: pick rxbar chocolate. Use 2D points to mark the target location where the object you need to manipulate in the task should ultimately be moved. The answer should be presented in JS...
er1_with_xml
[ "images/datasets/er1/images/FSD_points/fsd_pointrtx_19318/image_0.jpg" ]
[ "<tag_name name=\"rxbar chocolate\" img_idx=\"0\" point=\"[640, 431]\">" ]
"region_ref/1241161411-d5b363c_cam08_obj1-oven-drawer_0_2_board_1_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area delineated by a red rectangle. Locate a few points within an unoccupied area on the outlined surface. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1241161411-d5b363c_cam08_obj1-oven-drawer_0_2_board_1_on/image_0.png" ]
[ "<tag_name name=\"red rectangle\" img_idx=\"0\" point=\"[785, 916]\">" ]
"refcoco_19745"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: giaraafe standing tall. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the <answ...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_19745/image_0.jpg" ]
[ "<tag_name name=\"giraffe\" img_idx=\"0\" point=\"[671, 450]\">" ]
"roborefit_20789"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: can you find the big brown toy lion. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer betw...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_20789/image_0.png" ]
[ "<tag_name name=\"big brown toy lion\" img_idx=\"0\" point=\"[270, 783]\">" ]
"roborefit_27453"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: I need to get the boxed milk. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between the...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_27453/image_0.png" ]
[ "<tag_name name=\"boxed milk\" img_idx=\"0\" point=\"[468, 511]\">" ]
"refcoco_17329"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: A man in the background with a black satchel.. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final a...
er1_with_xml
[ "images/datasets/er1/images/refcoco/refcoco_17329/image_0.jpg" ]
[ "<tag_name name=\"man\" img_idx=\"0\" point=\"[205, 478]\">", "<tag_name name=\"satchel\" img_idx=\"0\" point=\"[179, 468]\">" ]
"object_ref/1310303335-502d64c_cam02_obj5-obj21_beside"
null
[ { "from": "human", "value": "<image>\nThe image features an item encircled by a red rectangle. Locate several spots on a different item that is next to the item marked by the rectangle. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and th...
er1_with_xml
[ "images/datasets/er1/images/robopoint/object_ref/1310303335-502d64c_cam02_obj5-obj21_beside/image_0.png" ]
[ "<tag_name name=\"vase/container\" img_idx=\"0\" point=\"[739, 548]\">", "<tag_name name=\"stone tool/anvil\" img_idx=\"0\" point=\"[601, 580]\">" ]
"roborefit_9783"
null
[ { "from": "human", "value": "<image>\nProvide one or more points coordinate of objects region this sentence describes: fetch the light brown toy lion. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer between t...
er1_with_xml
[ "images/datasets/er1/images/roborefit/roborefit_9783/image_0.png" ]
[ "<tag_name name=\"light brown toy lion\" img_idx=\"0\" point=\"[698, 574]\">" ]
"region_ref/1045764965-f53019d_cam10_obj18-oven-door_0_1_door_on"
null
[ { "from": "human", "value": "<image>\nThe image features an area outlined by a red rectangle. Locate several points in the vacant space on the delineated area. The answer should be presented in JSON format. Provide your thinking process between the <think> and </think> tags, and then give your final answer ...
er1_with_xml
[ "images/datasets/er1/images/robopoint/region_ref/1045764965-f53019d_cam10_obj18-oven-door_0_1_door_on/image_0.png" ]
[ "<tag_name name=\"drawer\" img_idx=\"0\" point=\"[739, 640]\">" ]