From 42fe741ff18381df2fc00b665f0b4585ab065fd7 Mon Sep 17 00:00:00 2001 From: tink2123 Date: Fri, 29 Jan 2021 15:08:58 +0800 Subject: [PATCH] add srn doc --- doc/doc_ch/algorithm_overview.md | 4 ++-- doc/doc_ch/inference.md | 21 +++++++++++++++++---- doc/doc_ch/recognition.md | 2 ++ doc/doc_en/algorithm_overview_en.md | 4 ++-- doc/doc_en/inference_en.md | 20 ++++++++++++++++++-- doc/doc_en/recognition_en.md | 1 + 6 files changed, 42 insertions(+), 10 deletions(-) diff --git a/doc/doc_ch/algorithm_overview.md b/doc/doc_ch/algorithm_overview.md index 59d1bc8c..f0765695 100755 --- a/doc/doc_ch/algorithm_overview.md +++ b/doc/doc_ch/algorithm_overview.md @@ -41,7 +41,7 @@ PaddleOCR基于动态图开源的文本识别算法列表： - [x] Rosetta([paper](https://arxiv.org/abs/1910.05085))[10] - [x] STAR-Net([paper](http://www.bmva.org/bmvc/2016/papers/paper043/index.html))[11] - [ ] RARE([paper](https://arxiv.org/abs/1603.03915v1))[12] coming soon -- [ ] SRN([paper](https://arxiv.org/abs/2003.12294))[5] coming soon +- [x] SRN([paper](https://arxiv.org/abs/2003.12294))[5] 参考[DTRB][3](https://arxiv.org/abs/1904.01906)文字识别训练和评估流程，使用MJSynth和SynthText两个文字识别数据集训练，在IIIT, SVT, IC03, IC13, IC15, SVTP, CUTE数据集上进行评估，算法效果如下： @@ -53,5 +53,5 @@ PaddleOCR基于动态图开源的文本识别算法列表： |CRNN|MobileNetV3|79.97%|rec_mv3_none_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_bilstm_ctc_v2.0_train.tar)| |StarNet|Resnet34_vd|84.44%|rec_r34_vd_tps_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_ctc_v2.0_train.tar)| |StarNet|MobileNetV3|81.42%|rec_mv3_tps_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_ctc_v2.0_train.tar)| - +|SRN|Resnet50_vd_fpn| 88.52% | rec_r50fpn_vd_none_srn | [下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r50_vd_srn_train.tar) | PaddleOCR文本识别算法的训练和使用请参考文档教程中[模型训练/评估中的文本识别部分](./recognition.md)。 diff --git a/doc/doc_ch/inference.md b/doc/doc_ch/inference.md index c4601e15..0daddd9b 100755 --- a/doc/doc_ch/inference.md +++ b/doc/doc_ch/inference.md @@ -22,8 +22,9 @@ inference 模型（`paddle.jit.save`保存的模型） - [三、文本识别模型推理](#文本识别模型推理) - [1. 超轻量中文识别模型推理](#超轻量中文识别模型推理) - [2. 基于CTC损失的识别模型推理](#基于CTC损失的识别模型推理) - - [3. 自定义文本识别字典的推理](#自定义文本识别字典的推理) - - [4. 多语言模型的推理](#多语言模型的推理) + - [3. 基于SRN损失的识别模型推理](#基于SRN损失的识别模型推理) + - [4. 自定义文本识别字典的推理](#自定义文本识别字典的推理) + - [5. 多语言模型的推理](#多语言模型的推理) - [四、方向分类模型推理](#方向识别模型推理) - [1. 方向分类模型推理](#方向分类模型推理) @@ -295,8 +296,20 @@ Predicts of ./doc/imgs_words_en/word_336.png:('super', 0.9999073) self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz" dict_character = list(self.character_str) ``` + +### 3. 基于SRN损失的识别模型推理 +基于SRN损失的识别模型，需要额外设置识别算法参数 --rec_algorithm="SRN"。 +同时需要保证预测shape与训练时一致，如： --rec_image_shape="1, 64, 256" -### 3. 自定义文本识别字典的推理 +``` +python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" \ + --rec_model_dir="./inference/srn/" \ + --rec_image_shape="1, 64, 256" \ + --rec_char_type="en" \ + --rec_algorithm="SRN" +``` + +### 4. 自定义文本识别字典的推理 如果训练时修改了文本的字典，在使用inference模型预测时，需要通过`--rec_char_dict_path`指定使用的字典路径，并且设置 `rec_char_type=ch` ``` @@ -304,7 +317,7 @@ python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png ``` -### 4. 多语言模型的推理 +### 5. 多语言模型的推理 如果您需要预测的是其他语言模型，在使用inference模型预测时，需要通过`--rec_char_dict_path`指定使用的字典路径, 同时为了得到正确的可视化结果， 需要通过 `--vis_font_path` 指定可视化的字体路径，`doc/fonts/` 路径下有默认提供的小语种字体，例如韩文识别： diff --git a/doc/doc_ch/recognition.md b/doc/doc_ch/recognition.md index c5f459bd..bc877ab7 100644 --- a/doc/doc_ch/recognition.md +++ b/doc/doc_ch/recognition.md @@ -36,6 +36,7 @@ ln -sf /train_data/dataset * 数据下载 若您本地没有数据集，可以在官网下载 [icdar2015](http://rrc.cvc.uab.es/?ch=4&com=downloads) 数据，用于快速验证。也可以参考[DTRB](https://github.com/clovaai/deep-text-recognition-benchmark#download-lmdb-dataset-for-traininig-and-evaluation-from-here)，下载 benchmark 所需的lmdb格式数据集。 +如果希望复现SRN的论文指标，需要下载离线[增广数据](https://pan.baidu.com/s/1-HSZ-ZVdqBF2HaBZ5pRAKA),提取码: y3ry。增广数据是由MJSynth和SynthText做旋转和扰动得到的。数据下载完成后请解压到 {your_path}/PaddleOCR/train_data/data_lmdb_release/training/ 路径下。 * 使用自己数据集 @@ -200,6 +201,7 @@ PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_t | rec_mv3_none_none_ctc.yml | Rosetta | Mobilenet_v3 large 0.5 | None | None | ctc | | rec_r34_vd_none_bilstm_ctc.yml | CRNN | Resnet34_vd | None | BiLSTM | ctc | | rec_r34_vd_none_none_ctc.yml | Rosetta | Resnet34_vd | None | None | ctc | +| rec_r50fpn_vd_none_srn.yml | SRN | Resnet50_fpn_vd | None | rnn | srn | 训练中文数据，推荐使用[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)，如您希望尝试其他算法在中文数据集上的效果，请参考下列说明修改配置文件： diff --git a/doc/doc_en/algorithm_overview_en.md b/doc/doc_en/algorithm_overview_en.md index 68bfd529..5016223f 100755 --- a/doc/doc_en/algorithm_overview_en.md +++ b/doc/doc_en/algorithm_overview_en.md @@ -43,7 +43,7 @@ PaddleOCR open-source text recognition algorithms list: - [x] Rosetta([paper](https://arxiv.org/abs/1910.05085))[10] - [x] STAR-Net([paper](http://www.bmva.org/bmvc/2016/papers/paper043/index.html))[11] - [ ] RARE([paper](https://arxiv.org/abs/1603.03915v1))[12] coming soon -- [ ] SRN([paper](https://arxiv.org/abs/2003.12294))[5] coming soon +- [x] SRN([paper](https://arxiv.org/abs/2003.12294))[5] Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation result of these above text recognition (using MJSynth and SynthText for training, evaluate on IIIT, SVT, IC03, IC13, IC15, SVTP, CUTE) is as follow: @@ -55,5 +55,5 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r |CRNN|MobileNetV3|79.97%|rec_mv3_none_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_bilstm_ctc_v2.0_train.tar)| |StarNet|Resnet34_vd|84.44%|rec_r34_vd_tps_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_ctc_v2.0_train.tar)| |StarNet|MobileNetV3|81.42%|rec_mv3_tps_bilstm_ctc|[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_ctc_v2.0_train.tar)| - +|SRN|Resnet50_vd_fpn| 88.52% | rec_r50fpn_vd_none_srn |[Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r50_vd_srn_train.tar)| Please refer to the document for training guide and use of PaddleOCR text recognition algorithms [Text recognition model training/evaluation/prediction](./recognition_en.md) diff --git a/doc/doc_en/inference_en.md b/doc/doc_en/inference_en.md index ccbb7184..c8ce1424 100755 --- a/doc/doc_en/inference_en.md +++ b/doc/doc_en/inference_en.md @@ -25,6 +25,7 @@ Next, we first introduce how to convert a trained model into an inference model, - [TEXT RECOGNITION MODEL INFERENCE](#RECOGNITION_MODEL_INFERENCE) - [1. LIGHTWEIGHT CHINESE MODEL](#LIGHTWEIGHT_RECOGNITION) - [2. CTC-BASED TEXT RECOGNITION MODEL INFERENCE](#CTC-BASED_RECOGNITION) + - [3. SRN-BASED TEXT RECOGNITION MODEL INFERENCE](#SRN-BASED_RECOGNITION) - [3. TEXT RECOGNITION MODEL INFERENCE USING CUSTOM CHARACTERS DICTIONARY](#USING_CUSTOM_CHARACTERS) - [4. MULTILINGUAL MODEL INFERENCE](MULTILINGUAL_MODEL_INFERENCE) @@ -304,8 +305,23 @@ self.character_str = "0123456789abcdefghijklmnopqrstuvwxyz" dict_character = list(self.character_str) ``` + +### 3. SRN-BASED TEXT RECOGNITION MODEL INFERENCE + +The recognition model based on SRN requires additional setting of the recognition algorithm parameter +--rec_algorithm="SRN". At the same time, it is necessary to ensure that the predicted shape is consistent +with the training, such as: --rec_image_shape="1, 64, 256" + +``` +python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png" \ + --rec_model_dir="./inference/srn/" \ + --rec_image_shape="1, 64, 256" \ + --rec_char_type="en" \ + --rec_algorithm="SRN" +``` + -### 3. TEXT RECOGNITION MODEL INFERENCE USING CUSTOM CHARACTERS DICTIONARY +### 4. TEXT RECOGNITION MODEL INFERENCE USING CUSTOM CHARACTERS DICTIONARY If the text dictionary is modified during training, when using the inference model to predict, you need to specify the dictionary path used by `--rec_char_dict_path`, and set `rec_char_type=ch` ``` @@ -313,7 +329,7 @@ python3 tools/infer/predict_rec.py --image_dir="./doc/imgs_words_en/word_336.png ``` -### 4. MULTILINGAUL MODEL INFERENCE +### 5. MULTILINGAUL MODEL INFERENCE If you need to predict other language models, when using inference model prediction, you need to specify the dictionary path used by `--rec_char_dict_path`. At the same time, in order to get the correct visualization results, You need to specify the visual font path through `--vis_font_path`. There are small language fonts provided by default under the `doc/fonts` path, such as Korean recognition: diff --git a/doc/doc_en/recognition_en.md b/doc/doc_en/recognition_en.md index 22f89cde..f29703d1 100644 --- a/doc/doc_en/recognition_en.md +++ b/doc/doc_en/recognition_en.md @@ -195,6 +195,7 @@ If the evaluation set is large, the test will be time-consuming. It is recommend | rec_mv3_none_none_ctc.yml | Rosetta | Mobilenet_v3 large 0.5 | None | None | ctc | | rec_r34_vd_none_bilstm_ctc.yml | CRNN | Resnet34_vd | None | BiLSTM | ctc | | rec_r34_vd_none_none_ctc.yml | Rosetta | Resnet34_vd | None | None | ctc | +| rec_r50fpn_vd_none_srn.yml | SRN | Resnet50_fpn_vd | None | rnn | srn | For training Chinese data, it is recommended to use [rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml). If you want to try the result of other algorithms on the Chinese data set, please refer to the following instructions to modify the configuration file: -- GitLab