update

README.md

@@ -11,29 +11,48 @@ tags:
 - onnx
 ---
 
-# ONNX - PP-DocLayout_plus-L
-FP32 = pp-docLayout_plus-l.onnx
+# ONNX - Paddle
+
+| Model | FP32 | Task |
+| --- | --- | --- |
+| PP-DocLayout_plus-L | pp-docLayout_plus-l.onnx | layout detection |
+| PP-OCRv6_medium_det | pp-ocrV6_medium_det.onnx | text line detection |
+| PP-OCRv6_medium_rec | pp-ocrV6_medium_rec.onnx | text recognition |
 
 ## Usage
 
 ```
 # onnxruntime-gpu for run it on GPU
-pip install onnxruntime opencv-python numpy
+pip install onnxruntime opencv-python numpy pyclipper
 ```
 
-- [PP-DocLayout_plus-L/src/helper.py](PP-DocLayout_plus-L/src/helper.py) = onnx session builder (provider selection, threads, memory options)
-- [PP-DocLayout_plus-L/src/example.py](PP-DocLayout_plus-L/src/example.py) = full pipeline: image preprocess, inference, box filter, label map
+Every model folder has the same layout:
+
+- `src/helper.py` = onnx session builder (provider selection, threads, memory options)
+- `src/example.py` = full pipeline: image preprocess, inference, postprocess
+- `onnx/` = the model file
+
+Run the commands from the repository root.
+
+Test images used by the examples:
+
+- [jp_1.jpg](https://github.com/cimo/One_workspace/blob/main/file/test/jp_1.jpg) = full page
+- [det_box_0294.jpg](https://github.com/cimo/One_workspace/blob/main/file/test/det_box_0294.jpg) = single text line
+
+## PP-DocLayout_plus-L
+
+- [PP-DocLayout_plus-L/src/example.py](PP-DocLayout_plus-L/src/example.py)
 
 ```
-python3 PP-DocLayout_plus-L/src/example.py page.jpg
+python3 PP-DocLayout_plus-L/src/example.py jp_1.jpg
 ```
 
 ```
-0.985695 | text | [79, 395, 819, 568]
-0.945875 | image | [858, 511, 1128, 774]
-0.944402 | paragraph_title | [81, 1472, 367, 1491]
-0.835724 | header | [109, 88, 256, 130]
-0.591139 | table | [837, 823, 1154, 1616]
+0.936555 | header | [530, 62, 1213, 104]
+0.910084 | table | [1066, 176, 1692, 260]
+0.891657 | footer | [1066, 1157, 1198, 1175]
+0.886924 | table | [54, 395, 1694, 799]
+0.827543 | text | [55, 804, 1029, 959]
 ```
 
 Note:
@@ -45,3 +64,52 @@ the box count is in the second output.
 full map in [PP-DocLayout_plus-L/src/example.py](PP-DocLayout_plus-L/src/example.py).
 - Boxes overlap by design (no NMS in the model): filter by score (0.3 in the example) and handle<br>
 the containment on your side if needed.
+
+## PP-OCRv6_medium_det
+
+- [PP-OCRv6_medium_det/src/example.py](PP-OCRv6_medium_det/src/example.py)
+
+```
+python3 PP-OCRv6_medium_det/src/example.py jp_1.jpg
+```
+
+```
+0.965487 | [[60, 1158], [850, 1158], [850, 1176], [60, 1176]]
+0.827508 | [[1069, 1156], [1200, 1156], [1200, 1176], [1069, 1176]]
+0.916008 | [[64, 1128], [837, 1128], [837, 1149], [64, 1149]]
+```
+
+Note:
+- Input = BGR image (plain `cv2.imread`, no channel swap) normalized with the ImageNet mean / std,<br>
+CHW with batch dimension.
+- Feed = `x`, the only input, with a fully dynamic shape.
+- Resize = longest side capped to 960 (4000 hard limit), then both sides rounded to a multiple of 32.
+- Output = a single probability map `[1, 1, height, width]`, the DB postprocess is on your side:<br>
+binarize at 0.2, `cv2.findContours`, minimum area box, score as the mean probability inside the box,<br>
+keep it over 0.45, expand it with pyclipper (unclip ratio 1.4) and rescale to the original image.
+- Output row = `[score, [4 corner points]]` clockwise from the top-left corner, already in the original<br>
+image space. Boxes are quads, not axis aligned: a rotated line keeps its rotation.
+- Crop the quads and feed them to PP-OCRv6_medium_rec to get the text.
+
+## PP-OCRv6_medium_rec
+
+- [PP-OCRv6_medium_rec/src/example.py](PP-OCRv6_medium_rec/src/example.py)
+
+```
+python3 PP-OCRv6_medium_rec/src/example.py det_box_0294.jpg
+```
+
+```
+0.953031 | (別表十(一)「15」若しくは別表十(二)「10」又は別表十(一)「16」若しくは別表十(二)「11」)
+```
+
+Note:
+- Input = a single cropped text line, not a full page: use the boxes from PP-OCRv6_medium_det.
+- Input = BGR image (plain `cv2.imread`, no channel swap) resized to height 48 keeping the aspect ratio,<br>
+scaled to 0..1 then normalized to -1..1, right padded with zeros, CHW with batch dimension.
+- Feed = `x`, the only input, with a dynamic width (minimum 320, capped at 3200).
+- Output = `[1, timeStep, 18710]` already softmaxed, decoded with a greedy CTC: drop the repeats first,<br>
+then the blank at index 0.
+- Character map = index 0 is the blank, 1..18708 are the lines of<br>
+[PP-OCRv6_medium_rec/onnx/dictionary.txt](PP-OCRv6_medium_rec/onnx/dictionary.txt), 18709 is the space.
+- Score = mean probability of the kept time steps.