Skip to content

Commit 2dcc6a9

Browse files
authoredSep 8, 2021
Merge pull request PaddlePaddle#3970 from Evezerest/2.3
Add dictionary, add_new_algorithm and doc link
2 parents 01853a4 + c6bd5ed commit 2dcc6a9

11 files changed

+87
-49
lines changed
 

‎README.md

+2-1
Original file line numberDiff line numberDiff line change
@@ -120,14 +120,15 @@ For a new language request, please refer to [Guideline for new language_requests
120120
- [Two-stage Algorithm](./doc/doc_en/algorithm_overview_en.md)
121121
- [PGNet Algorithm](./doc/doc_en/algorithm_overview_en.md)
122122
- [Python Inference](./doc/doc_en/inference_en.md)
123+
- [Use PaddleOCR Architecture to Add New Algorithms](./doc/doc_en/add_new_algorithm_en.md)
123124
- Data Annotation and Synthesis
124125
- [Semi-automatic Annotation Tool: PPOCRLabel](./PPOCRLabel/README.md)
125126
- [Data Synthesis Tool: Style-Text](./StyleText/README.md)
126127
- [Other Data Annotation Tools](./doc/doc_en/data_annotation_en.md)
127128
- [Other Data Synthesis Tools](./doc/doc_en/data_synthesis_en.md)
128129
- Datasets
129130
- [General OCR Datasets(Chinese/English)](./doc/doc_en/datasets_en.md)
130-
- [HandWritten_OCR_Datasets(Chinese)](./doc/doc_en/handwritten_datasets_en.md)
131+
- [Handwritten OCR Datasets(Chinese)](./doc/doc_en/handwritten_datasets_en.md)
131132
- [Various OCR Datasets(multilingual)](./doc/doc_en/vertical_and_multilingual_datasets_en.md)
132133
- [Visualization](#Visualization)
133134
- [New language requests](#language_requests)

‎README_ch.md

+3-2
Original file line numberDiff line numberDiff line change
@@ -94,7 +94,7 @@ PaddleOCR旨在打造一套丰富、领先、且实用的OCR工具库,助力
9494
- PP-OCR产业落地:从训练到部署
9595
- [PP-OCR模型库](./doc/doc_ch/models.md)
9696
- [PP-OCR模型下载](./doc/doc_ch/models_list.md)
97-
- [PP-OCR模型库Python推理](./doc/doc_ch/inference_ppocr.md)
97+
- [基于Python引擎的PP-OCR模型库推理](./doc/doc_ch/inference_ppocr.md)
9898
- [PP-OCR模型训练](./doc/doc_ch/training.md)
9999
- [文本检测](./doc/doc_ch/detection.md)
100100
- [文本识别](./doc/doc_ch/recognition.md)
@@ -114,9 +114,10 @@ PaddleOCR旨在打造一套丰富、领先、且实用的OCR工具库,助力
114114
- [其它数据标注工具](./doc/doc_ch/data_annotation.md)
115115
- [其它数据合成工具](./doc/doc_ch/data_synthesis.md)
116116
- OCR学术圈
117-
- [两阶段模型介绍与下载](./doc/doc_ch/algorithm_overview.md)
117+
- [两阶段算法](./doc/doc_ch/algorithm_overview.md)
118118
- [端到端PGNet算法](./doc/doc_ch/pgnet.md)
119119
- [基于Python脚本预测引擎推理](./doc/doc_ch/inference.md)
120+
- [使用PaddleOCR架构添加新算法](./doc/doc_ch/add_new_algorithm.md)
120121
- 数据集
121122
- [通用中英文OCR数据集](./doc/doc_ch/datasets.md)
122123
- [手写中文OCR数据集](./doc/doc_ch/handwritten_datasets.md)

‎doc/doc_ch/algorithm_overview.md

+45-29
Original file line numberDiff line numberDiff line change
@@ -1,12 +1,20 @@
1-
<a name="算法介绍"></a>
2-
## 算法介绍
1+
# 两阶段算法
2+
3+
- [两阶段算法](#-----)
4+
* [1. 算法介绍](#1)
5+
+ [1.1 文本检测算法](#11)
6+
+ [1.2 文本识别算法](#12)
7+
* [2. 模型训练](#2)
8+
* [3. 模型推理](#3)
9+
10+
<a name="1"></a>
11+
12+
## 1. 算法介绍
313
本文给出了PaddleOCR已支持的文本检测算法和文本识别算法列表,以及每个算法在**英文公开数据集**上的模型和指标,主要用于算法简介和算法性能对比,更多包括中文在内的其他数据集上的模型请参考[PP-OCR v2.0 系列模型下载](./models_list.md)
414

5-
- [1.文本检测算法](#文本检测算法)
6-
- [2.文本识别算法](#文本识别算法)
15+
<a name="11"></a>
716

8-
<a name="文本检测算法"></a>
9-
### 1.文本检测算法
17+
### 1.1 文本检测算法
1018

1119
PaddleOCR开源的文本检测算法列表:
1220
- [x] DB([paper]( https://arxiv.org/abs/1911.08947)) [2](ppocr推荐)
@@ -16,27 +24,25 @@ PaddleOCR开源的文本检测算法列表:
1624
在ICDAR2015文本检测公开数据集上,算法效果如下:
1725
|模型|骨干网络|precision|recall|Hmean|下载链接|
1826
| --- | --- | --- | --- | --- | --- |
19-
|EAST|ResNet50_vd|85.80%|86.71%|86.25%|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_east_v2.0_train.tar)|
20-
|EAST|MobileNetV3|79.42%|80.64%|80.03%|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_east_v2.0_train.tar)|
21-
|DB|ResNet50_vd|86.41%|78.72%|82.38%|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_db_v2.0_train.tar)|
22-
|DB|MobileNetV3|77.29%|73.08%|75.12%|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar)|
23-
|SAST|ResNet50_vd|91.39%|83.77%|87.42%|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_icdar15_v2.0_train.tar)|
27+
|EAST|ResNet50_vd|85.80%|86.71%|86.25%|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_east_v2.0_train.tar)|
28+
|EAST|MobileNetV3|79.42%|80.64%|80.03%|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_east_v2.0_train.tar)|
29+
|DB|ResNet50_vd|86.41%|78.72%|82.38%|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_db_v2.0_train.tar)|
30+
|DB|MobileNetV3|77.29%|73.08%|75.12%|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_mv3_db_v2.0_train.tar)|
31+
|SAST|ResNet50_vd|91.39%|83.77%|87.42%|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_icdar15_v2.0_train.tar)|
2432

2533
在Total-text文本检测公开数据集上,算法效果如下:
2634

2735
|模型|骨干网络|precision|recall|Hmean|下载链接|
2836
| --- | --- | --- | --- | --- | --- |
29-
|SAST|ResNet50_vd|89.63%|78.44%|83.66%|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_totaltext_v2.0_train.tar)|
37+
|SAST|ResNet50_vd|89.63%|78.44%|83.66%|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/det_r50_vd_sast_totaltext_v2.0_train.tar)|
3038

3139
**说明:** SAST模型训练额外加入了icdar2013、icdar2017、COCO-Text、ArT等公开数据集进行调优。PaddleOCR用到的经过整理格式的英文公开数据集下载:
3240
* [百度云地址](https://pan.baidu.com/s/12cPnZcVuV1zn5DOd4mqjVw) (提取码: 2bpi)
3341
* [Google Drive下载地址](https://drive.google.com/drive/folders/1ll2-XEVyCQLpJjawLDiRlvo_i4BqHCJe?usp=sharing)
3442

35-
PaddleOCR文本检测算法的训练和使用请参考文档教程中[模型训练/评估中的文本检测部分](./detection.md)
43+
<a name="12"></a>
3644

37-
38-
<a name="文本识别算法"></a>
39-
### 2.文本识别算法
45+
### 1.2 文本识别算法
4046

4147
PaddleOCR基于动态图开源的文本识别算法列表:
4248
- [x] CRNN([paper](https://arxiv.org/abs/1507.05717))[7](ppocr推荐)
@@ -50,16 +56,26 @@ PaddleOCR基于动态图开源的文本识别算法列表:
5056

5157
|模型|骨干网络|Avg Accuracy|模型存储命名|下载链接|
5258
|---|---|---|---|---|
53-
|Rosetta|Resnet34_vd|80.9%|rec_r34_vd_none_none_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_none_ctc_v2.0_train.tar)|
54-
|Rosetta|MobileNetV3|78.05%|rec_mv3_none_none_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_none_ctc_v2.0_train.tar)|
55-
|CRNN|Resnet34_vd|82.76%|rec_r34_vd_none_bilstm_ctc|[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_bilstm_ctc_v2.0_train.tar)|
56-
|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)|
57-
|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)|
58-
|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)|
59-
|RARE|MobileNetV3|82.5%|rec_mv3_tps_bilstm_att |[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_att_v2.0_train.tar)|
60-
|RARE|Resnet34_vd|83.6%|rec_r34_vd_tps_bilstm_att |[下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_att_v2.0_train.tar)|
61-
|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) |
62-
|NRTR|NRTR_MTB| 84.3% | rec_mtb_nrtr | [下载链接](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mtb_nrtr_train.tar) |
63-
64-
65-
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)
59+
|Rosetta|Resnet34_vd|80.9%|rec_r34_vd_none_none_ctc|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_none_ctc_v2.0_train.tar)|
60+
|Rosetta|MobileNetV3|78.05%|rec_mv3_none_none_ctc|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_none_ctc_v2.0_train.tar)|
61+
|CRNN|Resnet34_vd|82.76%|rec_r34_vd_none_bilstm_ctc|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_none_bilstm_ctc_v2.0_train.tar)|
62+
|CRNN|MobileNetV3|79.97%|rec_mv3_none_bilstm_ctc|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_none_bilstm_ctc_v2.0_train.tar)|
63+
|StarNet|Resnet34_vd|84.44%|rec_r34_vd_tps_bilstm_ctc|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_ctc_v2.0_train.tar)|
64+
|StarNet|MobileNetV3|81.42%|rec_mv3_tps_bilstm_ctc|[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_ctc_v2.0_train.tar)|
65+
|RARE|MobileNetV3|82.5%|rec_mv3_tps_bilstm_att |[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mv3_tps_bilstm_att_v2.0_train.tar)|
66+
|RARE|Resnet34_vd|83.6%|rec_r34_vd_tps_bilstm_att |[预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r34_vd_tps_bilstm_att_v2.0_train.tar)|
67+
|SRN|Resnet50_vd_fpn| 88.52% | rec_r50fpn_vd_none_srn | [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_r50_vd_srn_train.tar) |
68+
|NRTR|NRTR_MTB| 84.3% | rec_mtb_nrtr | [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mtb_nrtr_train.tar) |
69+
70+
<a name="2"></a>
71+
72+
## 2. 模型训练
73+
74+
PaddleOCR文本检测算法的训练和使用请参考文档教程中[模型训练/评估中的文本检测部分](./detection.md)。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)
75+
76+
<a name="3"></a>
77+
78+
## 3. 模型推理
79+
80+
上述模型中除PP-OCR系列模型以外,其余模型仅支持基于Python引擎的推理,具体内容可参考[基于Python预测引擎推理](./inference.md)
81+

‎doc/doc_ch/inference_ppocr.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
# PP-OCR模型库Python推理
1+
# 基于Python引擎的PP-OCR模型库推理
22

33
本文介绍针对PP-OCR模型库的Python推理引擎使用方法,内容依次为文本检测、文本识别、方向分类器以及三者串联在CPU、GPU上的预测方法。
44

‎doc/doc_ch/models_list.md

+4-4
Original file line numberDiff line numberDiff line change
@@ -33,8 +33,8 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训
3333

3434
|模型名称|模型简介|配置文件|推理模型大小|下载地址|
3535
| --- | --- | --- | --- | --- |
36-
|ch_PP-OCRv2_det_slim|slim量化+蒸馏版超轻量模型,支持中英文、多语种文本检测|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)| 3M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar)|
37-
|ch_PP-OCRv2_det|原始超轻量模型,支持中英文、多语种文本检测|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)|3M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)|
36+
|ch_PP-OCRv2_det_slim|【最新】slim量化+蒸馏版超轻量模型,支持中英文、多语种文本检测|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)| 3M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar)|
37+
|ch_PP-OCRv2_det|【最新】原始超轻量模型,支持中英文、多语种文本检测|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)|3M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)|
3838
|ch_ppocr_mobile_slim_v2.0_det|slim裁剪版超轻量模型,支持中英文、多语种文本检测|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)| 2.6M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_prune_infer.tar)|
3939
|ch_ppocr_mobile_v2.0_det|原始超轻量模型,支持中英文、多语种文本检测|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)|3M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|
4040
|ch_ppocr_server_v2.0_det|通用模型,支持中英文、多语种文本检测,比超轻量模型更大,但效果更好|[ch_det_res18_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_res18_db_v2.0.yml)|47M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)|
@@ -48,8 +48,8 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训
4848

4949
|模型名称|模型简介|配置文件|推理模型大小|下载地址|
5050
| --- | --- | --- | --- | --- |
51-
|ch_PP-OCRv2_rec_slim|slim量化版超轻量模型,支持中英文、数字识别|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) |
52-
|ch_PP-OCRv2_rec|原始超轻量模型,支持中英文、数字识别|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)|8.5M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
51+
|ch_PP-OCRv2_rec_slim|【最新】slim量化版超轻量模型,支持中英文、数字识别|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) |
52+
|ch_PP-OCRv2_rec|【最新】原始超轻量模型,支持中英文、数字识别|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)|8.5M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
5353
|ch_ppocr_mobile_slim_v2.0_rec|slim裁剪量化版超轻量模型,支持中英文、数字识别|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)| 6M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_train.tar) |
5454
|ch_ppocr_mobile_v2.0_rec|原始超轻量模型,支持中英文、数字识别|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)|5.2M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
5555
|ch_ppocr_server_v2.0_rec|通用模型,支持中英文、数字识别|[rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml)|94.8M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_train.tar) / [预训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |

‎doc/doc_ch/pgnet.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -43,7 +43,7 @@ PGNet算法细节详见[论文](https://www.aaai.org/AAAI21Papers/AAAI-2885.Wang
4343

4444
<a name="环境配置"></a>
4545
## 二、环境配置
46-
请先参考[快速安装](./installation.md)配置PaddleOCR运行环境
46+
请先参考[《运行环境准备》](./environment.md)配置PaddleOCR运行环境,参考[《PaddleOCR全景图与项目克隆》](./paddleOCR_overview.md)克隆项目
4747

4848
<a name="快速使用"></a>
4949
## 三、快速使用

‎doc/doc_en/add_new_algorithm_en.md

+2-2
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
# Add new algorithm
1+
# Add New Algorithm
22

33
PaddleOCR decomposes an algorithm into the following parts, and modularizes each part to make it more convenient to develop new algorithms.
44

@@ -263,7 +263,7 @@ Metric:
263263
main_indicator: acc
264264
```
265265
266-
## 优化器
266+
## Optimizer
267267
268268
The optimizer is used to train the network. The optimizer also contains network regularization and learning rate decay modules. This part is under [ppocr/optimizer](../../ppocr/optimizer). PaddleOCR has built-in
269269
Commonly used optimizer modules such as `Momentum`, `Adam` and `RMSProp`, common regularization modules such as `Linear`, `Cosine`, `Step` and `Piecewise`, and common learning rate decay modules such as `L1Decay` and `L2Decay`.

‎doc/doc_en/algorithm_overview_en.md

+22-4
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,14 @@
1+
# Two-stage Algorithm
2+
3+
- [1. Algorithm Introduction](#1-algorithm-introduction)
4+
* [1.1 Text Detection Algorithm](#11-text-detection-algorithm)
5+
* [1.2 Text Recognition Algorithm](#12-text-recognition-algorithm)
6+
- [2. Training](#2-training)
7+
- [3. Inference](#3-inference)
8+
19
<a name="Algorithm_introduction"></a>
2-
## Algorithm introduction
10+
11+
## 1. Algorithm Introduction
312

413
This tutorial lists the text detection algorithms and text recognition algorithms supported by PaddleOCR, as well as the models and metrics of each algorithm on **English public datasets**. It is mainly used for algorithm introduction and algorithm performance comparison. For more models on other datasets including Chinese, please refer to [PP-OCR v2.0 models list](./models_list_en.md).
514

@@ -8,7 +17,8 @@ This tutorial lists the text detection algorithms and text recognition algorithm
817
- [2. Text Recognition Algorithm](#TEXTRECOGNITIONALGORITHM)
918

1019
<a name="TEXTDETECTIONALGORITHM"></a>
11-
### 1. Text Detection Algorithm
20+
21+
### 1.1 Text Detection Algorithm
1222

1323
PaddleOCR open source text detection algorithms list:
1424
- [x] EAST([paper](https://arxiv.org/abs/1704.03155))[2]
@@ -38,7 +48,7 @@ On Total-Text dataset, the text detection result is as follows:
3848
For the training guide and use of PaddleOCR text detection algorithms, please refer to the document [Text detection model training/evaluation/prediction](./detection_en.md)
3949

4050
<a name="TEXTRECOGNITIONALGORITHM"></a>
41-
### 2. Text Recognition Algorithm
51+
### 1.2 Text Recognition Algorithm
4252

4353
PaddleOCR open-source text recognition algorithms list:
4454
- [x] CRNN([paper](https://arxiv.org/abs/1507.05717))[7]
@@ -63,4 +73,12 @@ Refer to [DTRB](https://arxiv.org/abs/1904.01906), the training and evaluation r
6373
|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)|
6474
|NRTR|NRTR_MTB| 84.3% | rec_mtb_nrtr | [Download link](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/rec_mtb_nrtr_train.tar) |
6575

66-
Please refer to the document for training guide and use of PaddleOCR text recognition algorithms [Text recognition model training/evaluation/prediction](./recognition_en.md)
76+
Please refer to the document for training guide and use of PaddleOCR
77+
78+
## 2. Training
79+
80+
For the training guide and use of PaddleOCR text detection algorithms, please refer to the document [Text detection model training/evaluation/prediction](./detection_en.md). For text recognition algorithms, please refer to [Text recognition model training/evaluation/prediction](./recognition_en.md)
81+
82+
## 3. Inference
83+
84+
Except for the PP-OCR series models of the above models, the other models only support inference based on the Python engine. For details, please refer to [Inference based on Python prediction engine](./inference_en.md)

‎doc/doc_en/environment_en.md

+2
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,7 @@
11
# Environment Preparation
22

3+
Windows and Mac users are recommended to use Anaconda to build a Python environment, and Linux users are recommended to use docker to build a Python environment. If you are familiar with the Python environment, you can skip to step 2 to install PaddlePaddle.
4+
35
* [1. Python Environment Setup](#1)
46
+ [1.1 Windows](#1.1)
57
+ [1.2 Mac](#1.2)

‎doc/doc_en/models_list_en.md

+4-4
Original file line numberDiff line numberDiff line change
@@ -29,8 +29,8 @@ Relationship of the above models is as follows.
2929

3030
|model name|description|config|model size|download|
3131
| --- | --- | --- | --- | --- |
32-
|ch_PP-OCRv2_det_slim|slim quantization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)| 3M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar)|
33-
|ch_PP-OCRv2_det|Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)|3M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)|
32+
|ch_PP-OCRv2_det_slim|[New] slim quantization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)| 3M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar)|
33+
|ch_PP-OCRv2_det|[New] Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCR_det_cml.yml)|3M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)|
3434
|ch_ppocr_mobile_slim_v2.0_det|Slim pruned lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)|2.6M |[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/slim/ch_ppocr_mobile_v2.0_det_prune_infer.tar)|
3535
|ch_ppocr_mobile_v2.0_det|Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_det_mv3_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml)|3M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar)|
3636
|ch_ppocr_server_v2.0_det|General model, which is larger than the lightweight model, but achieved better performance|[ch_det_res18_db_v2.0.yml](../../configs/det/ch_ppocr_v2.0/ch_det_res18_db_v2.0.yml)|47M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_det_train.tar)|
@@ -43,8 +43,8 @@ Relationship of the above models is as follows.
4343

4444
|model name|description|config|model size|download|
4545
| --- | --- | --- | --- | --- |
46-
|ch_PP-OCRv2_rec_slim|Slim qunatization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) |
47-
|ch_PP-OCRv2_rec|Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)|8.5M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
46+
|ch_PP-OCRv2_rec_slim|[New] Slim qunatization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) |
47+
|ch_PP-OCRv2_rec|[New] Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)|8.5M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) |
4848
|ch_ppocr_mobile_slim_v2.0_rec|Slim pruned and quantized lightweight model, supporting Chinese, English and number recognition|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)| 6M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_train.tar) |
4949
|ch_ppocr_mobile_v2.0_rec|Original lightweight model, supporting Chinese, English and number recognition|[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)|5.2M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_train.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_pre.tar) |
5050
|ch_ppocr_server_v2.0_rec|General model, supporting Chinese, English and number recognition|[rec_chinese_common_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_common_train_v2.0.yml)|94.8M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_train.tar) / [pre-trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_server_v2.0_rec_pre.tar) |

‎doc/doc_en/pgnet_en.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -36,7 +36,7 @@ The results of detection and recognition are as follows:
3636

3737
<a name="Environment_Configuration"></a>
3838
## 2. Environment Configuration
39-
Please refer to [Quick Installation](./installation_en.md) Configure the PaddleOCR running environment.
39+
Please refer to [Operation Environment Preparation](./environment_en.md) to configure PaddleOCR operating environment first, refer to [PaddleOCR Overview and Project Clone](./paddleOCR_overview_en.md) to clone the project
4040

4141
<a name="Quick_Use"></a>
4242
## 3. Quick Use

0 commit comments

Comments
 (0)
Please sign in to comment.