Add new recognition method "ParseQ" #10836
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* Update PP-OCRv4_introduction.md * Update PP-OCRv4_introduction.md * Update PP-OCRv4_introduction.md
…Paddle:Release/2.7 (PaddlePaddle#10655) * Don't break overall processing on a bad image * Add preprocessing common to OCR tasks Add preprocessing to options
added missing pyyaml library
Co-authored-by: Dennis <dvorst@users.noreply.github.com> Co-authored-by: shiyutang <34859558+shiyutang@users.noreply.github.com>
* 修改数据增强导致的DSR报错 * 错误修改回滚
Fixed simple spelling errors.
|
Thanks for your contribution! |
|
与Torch模型在Real(COCO-Text, RCTW17, Uber-Text, ArT, LSVT, MLT19, ReCTS, TextOCR and OpenVINO)数据集下训练，在IIIT5k, SVT, IC13,IC15,SVTP,CUTE数据集下测试结果对比： |
|
静态图模型导出成功： |
tink2123
reviewed
Sep 7, 2023
configs/rec/rec_vit_parseq.yml
Outdated
| use_visualdl: False | ||
| infer_img: doc/imgs_words_en/word_10.png | ||
| # for data or label process | ||
| character_dict_path: /ssd3/suzejia/newAlgOCR/ppocr/utils/dict/parseq_dict.txt |
There was a problem hiding this comment.
字典已上传，并相应修改configs中的路径
ppocr/data/imaug/rec_img_aug.py
Outdated
| self.image_shape = image_shape | ||
| self.dst_h, self.dst_w = image_shape[1], image_shape[2] | ||
|
|
||
| def __call__(self, data): |
There was a problem hiding this comment.
没有特殊的操作，可以看下能否复用之前的resize呢？
| ones_(m.weight) | ||
|
|
||
| def forward_features(self, x): | ||
| # B = x.shape[0] |
| from typing import Optional | ||
| import copy | ||
| from itertools import permutations | ||
|
|
There was a problem hiding this comment.
如果代码参考了其他repo，加一下 code refer, 参考：
This code is refer from:
https://github.com/PaddlePaddle/PaddleClas/blob/release%2F2.5/ppcls/arch/backbone/model_zoo/vision_transformer.py
| sz = perm.shape[0] | ||
| mask = paddle.zeros(shape=(sz, sz)) | ||
| for i in range(sz): | ||
| query_idx = perm[i].cpu().numpy().tolist() |
There was a problem hiding this comment.
嗯嗯，如果不转成list，直接用tensor的话，下面的mask索引会出bug
tools/infer/predict_rec.py
Outdated
| @@ -348,6 +354,17 @@ def resize_norm_img_svtr(self, img, image_shape): | |||
| resized_image /= 0.5 | |||
| return resized_image | |||
|
|
|||
| def resize_norm_img_parseq(self, img, image_shape): | |||
已修改至dygraph分支 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.