📄 Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.
add model converts to onnx format link
SWHL2ca6ef5e58dd
diff --git a/README.md b/README.mdindex 896cae1..0551409 100755--- a/README.md+++ b/README.md@@ -26,6 +26,8 @@- [项目结构](#项目结构)- [当前进展](#当前进展)- [模型相关](#模型相关)+ - [可以直接下载使用的模型 (下载链接:提取码:30jv)](#可以直接下载使用的模型-下载链接提取码30jv)+ - [模型转onnx](#模型转onnx)- [原始发起者及初创作者](#原始发起者及初创作者)- [版权声明](#版权声明)- [授权](#授权)@@ -50,6 +52,7 @@#### 🎵 2021-07-17 update- 完善README文档- 增加**英文、数字识别**onnx模型,具体参见`python/en_number_ppocr_mobile_v2_rec`,用法同其他+- 整理一下[模型转onnx](#模型转onnx)<details><summary>之前更新记录</summary>@@ -117,7 +120,6 @@├── ios # 苹果手机平台工程目录├── jvm # 基于java的工程目录├── lib # 编译用库文件目录,用于编译c语言接口库用,默认并不上传二进制文件- ├── models # 放置可使用的模型文件下载信息,基于百度网盘├── ocrweb # 基于python和Flask web├── python # python推理代码目录├── release #@@ -136,7 +138,7 @@- [ ] 依据python版本重写C++推理代码,以提升推理效果,并增加对gif/tga/webp 格式图片的支持## 模型相关-- 可以直接下载使用的模型 (下载链接:[提取码:30jv](https://pan.baidu.com/s/1qkqWK4wRdMjqGGbzR-FyWg))+#### 可以直接下载使用的模型 (下载链接:[提取码:30jv](https://pan.baidu.com/s/1qkqWK4wRdMjqGGbzR-FyWg))```textch_ppocr_mobile_v2.0_det_infer.onnxch_ppocr_mobile_v2.0_cls_infer.onnx@@ -146,10 +148,12 @@ch_ppocr_server_v2.0_rec_infer.onnxjapan_rec_crnn.onnx+ en_number_mobile_v2.0_rec_infer.onnx```--- [模型转换说明](./models)-- [手把手教你使用ONNXRunTime部署PP-OCR](https://aistudio.baidu.com/aistudio/projectdetail/1479970?channelType=0&channel=0) by @Channingss+#### 模型转onnx+ - [手把手教你使用ONNXRunTime部署PP-OCR](https://aistudio.baidu.com/aistudio/projectdetail/1479970?channelType=0&channel=0) by @Channingss+ - [✧✧PaddleOCRModelConverter](https://github.com/RapidAI/PaddleOCRModelConverter) by @[SWHL](https://github.com/SWHL)+ - [Paddle2OnnxConvertor](https://github.com/RapidAI/Paddle2OnnxConvertor) by @[benjaminwan](https://github.com/benjaminwan)## 原始发起者及初创作者diff --git a/README_en.md b/README_en.mdindex 1698d43..3621f59 100644--- a/README_en.md+++ b/README_en.md@@ -23,10 +23,10 @@ English | [Chinese](README.md)- [Online demo](#online-demo)- [Directory structure](#directory-structure)- [Current Progress](#current-progress)- - [Model Conversion](#model-conversion)- - [Model Conversion Instructions](#model-conversion-instructions)- - [PaddleOCR-FAQ](#paddleocr-faq)- - [ONNX model download](#onnx-model-download)+ - [Model related](#model-related)+ - [Download models](#download-models)+ - [Model to onnx](#model-to-onnx)+ - [PaddleOCR-FAQ](#paddleocr-faq)- [Original initiator and start-up author](#original-initiator-and-start-up-author)- [Authorization](#authorization)- [Contact us](#contact-us)@@ -50,6 +50,7 @@ English | [Chinese](README.md)#### 🎵 2021-07-17 update- Improve the README document- Add **English, number recognition**onnx model, please refer to `python/en_number_ppocr_mobile_v2_rec` for details, the usage is the same as others+- Organize [Model to onnx](#model-related)<details><summary>Previous update logs</summary>@@ -112,7 +113,6 @@ Since ubuntu users are all commercial users and have the ability to compile, pre├── ios # Apple mobile phone platform project directory├── jvm # java-based project directory├── lib # Compilation library file directory, used to compile the C language interface library. Binary files are not uploaded by default- ├── models # Place available model file download information, based on Baidu network disk├── ocrweb # Based on python and Flask web├── python # python reasoning code directory├── release #@@ -129,17 +129,26 @@ Since ubuntu users are all commercial users and have the ability to compile, pre- [ ] IOS example: waiting for someone to contribute code- [ ] Rewrite the C++ reasoning code according to the python version to improve the reasoning effect, and add support for gif/tga/webp format pictures-## Model Conversion-- The models currently supported by the conversion script:- - 1 text direction classification model,- - 2 detection models,- - 28 recognition models (2 in simplified Chinese, 26 in traditional Chinese, etc.), totaling 31-### [Model Conversion Instructions](./models)--### [PaddleOCR-FAQ](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/FAQ.md)--### ONNX model download-- [Extraction code: 30jv](https://pan.baidu.com/s/1qkqWK4wRdMjqGGbzR-FyWg)+## Model related+#### Download models+ - (download link: [extract code: 30jv](https://pan.baidu.com/s/1qkqWK4wRdMjqGGbzR-FyWg))+ ```text+ ch_ppocr_mobile_v2.0_det_infer.onnx+ ch_ppocr_mobile_v2.0_cls_infer.onnx+ ch_ppocr_mobile_v2.0_rec_infer.onnx++ ch_ppocr_server_v2.0_det_infer.onnx+ ch_ppocr_server_v2.0_rec_infer.onnx++ japan_rec_crnn.onnx+ en_number_mobile_v2.0_rec_infer.onnx+ ```+#### Model to onnx+ - [Teach you to use ONNXRunTime to deploy PP-OCR](https://aistudio.baidu.com/aistudio/projectdetail/1479970?channelType=0&channel=0) by @Channingss+ - [✧✧PaddleOCRModelConverter](https://github.com/RapidAI/PaddleOCRModelConverter) by @[SWHL](https://github.com/SWHL)+ - [Paddle2OnnxConvertor](https://github.com/RapidAI/Paddle2OnnxConvertor) by @[benjaminwan](https://github.com/benjaminwan)++## [PaddleOCR-FAQ](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/FAQ.md)## Original initiator and start-up author- [benjaminwan](https://github.com/benjaminwan)diff --git a/models/README.md b/models/README.mddeleted file mode 100755index 008baa7..0000000--- a/models/README.md+++ /dev/null@@ -1,301 +0,0 @@-## onnx模型下载-### 最新模型 (`2021-06-07 update`)-- 支持`onnxruntime>=1.7.0`,速度更快,内存占用更少-- 下载链接:[提取码:30jv](https://pan.baidu.com/s/1qkqWK4wRdMjqGGbzR-FyWg)--### 之前模型-- 只支持`onnxruntime==1.5.0`-- [点击下载](https://github.com/RapidOCR/RapidOCR/releases/download/V1.0/rapid-model.tgz)------## 2021-06-07 update: 后续会逐步更新这块,以下只是暂时的-## onnx模型转换说明-### 一、如何获得模型--* 您可以通过使用paddle全家桶,通过其完善的说明,事无巨细的指引,手把手地让您从零开始,训练自己的模型。-* 或者,您也可以直接去paddle官网下载预先训练好的模型。-* 这个预先训练好的模型叫作train模型,server模型体积较大(准确度高速度慢些),mobile模型体积较小(准确度相对低速度快)。-* det和rec有server和mobile两种模型,cls只有mobile模型。-* 预训练模型下载地址: [model_list](https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.1/doc/doc_ch/models_list.md)--* 使用`tools`下的`getmodels.sh`脚本下载并准备。--### 二、准备环境和工具--* 开始之前,需要先准备好Linux,paddle的pip包不支持macOS,而windows部署起来麻烦一些,以下说明以deepin20为范例。-* 安装python3: ```sudo apt-get install python3```-* 安装pip3:- ```shell- wget https://bootstrap.pypa.io/get-pip.py- sudo python3 get-pip.py- ```--* pip安装paddlepaddle( 2.0的模型需要2.0版):- ```shell- sudo python -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple- ```--- pip安装转换工具附加依赖:- ```shell- sudo pip3 install setuptools tqdm opencv-python imgaug pyclipper -i https://mirror.baidu.com/pypi/simple- ```--* 安装lmdb依赖:- ```shell- sudo apt-get install python3-yaml libxml2-dev libxslt1-dev python3-dev- ```-* pip安装lmdb:- ```shell- sudo pip3 install lmdb -i https://mirror.baidu.com/pypi/simple- ```-* pip安装onnx:- ```shell- sudo pip3 install onnx- ```-* 下载PaddleOCR(用于预训练模型转paddle推理模型):- ```shell- git clone https://github.com/PaddlePaddle/PaddleOCR.git- ```-* 安装转换工具库Paddle2ONNX(develop分支,用于paddle推理模型转onnx模型):- ```shell- # 转换工具修复,来自于官方仓库,包括未合并代码- git clone https://github.com/RapidOCR/paddle2onnx-wild.git- cd paddle2onnx-wild- sudo python3 setup.py install- ```--### 三、开始转换--* 接下来的操作都在PaddleOCR文件夹里进行-* PaddleOCR切换到release/2.0分支(因为train训练模型也是v2.0格式的)- ```shell- git clone https://github.com/PaddlePaddle/PaddleOCR.git- cd PaddleOCR- git checkout release/2.0- ```--* 生成多语言识别模型配置文件-- ```shell- #安装模块依赖- sudo pip3 install pyyaml- cd PaddleOCR/configs/rec/multi_language-- #生成多语言配置文件- python3 generate_multi_language_configs.py -l it- python3 generate_multi_language_configs.py -l es- python3 generate_multi_language_configs.py -l pt- python3 generate_multi_language_configs.py -l ru- python3 generate_multi_language_configs.py -l ar- python3 generate_multi_language_configs.py -l ta- python3 generate_multi_language_configs.py -l ug- python3 generate_multi_language_configs.py -l fa- python3 generate_multi_language_configs.py -l ur- python3 generate_multi_language_configs.py -l rs_latin- python3 generate_multi_language_configs.py -l oc- python3 generate_multi_language_configs.py -l rs_cyrillic- python3 generate_multi_language_configs.py -l bg- python3 generate_multi_language_configs.py -l uk- python3 generate_multi_language_configs.py -l be- python3 generate_multi_language_configs.py -l te- python3 generate_multi_language_configs.py -l kn- python3 generate_multi_language_configs.py -l ch_tra- python3 generate_multi_language_configs.py -l hi- python3 generate_multi_language_configs.py -l mr- python3 generate_multi_language_configs.py -l ne- ```--* 新建文件夹`PaddleOCR/inference`-* 需要确认系统中有没有`wget`工具,运行下`wget`检查确认- ```shell- # 在ubuntu下安装- $ apt install wget-- # 在centos安装:- $ yum install wget- ```--* 将本仓库`tools`目录下的`getmodels.sh` 放`PaddleOCR/inference`下--* 执行 chmod +x getmodels.h- 应该是如下结构:- ```text- PaddleOCR/inference- |- |-- getmodels.sh-- ```-* 运行`getmodels.sh`脚本: `./getmodels.sh`-* 运行完后将有如下的目录结构(大约是这样,有一些小的名字差异,懒得改了,不要太介意)- <details>- <summary>PaddleOCR/inference目录结构 (click to expand)</summary>-- ```text- PaddleOCR/inference- ├── ar_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── be_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── bg_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ch_ppocr_mobile_v2.0_cls_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ch_ppocr_mobile_v2.0_det_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ch_ppocr_mobile_v2.0_rec_train- │ ├── best_accuracy.pdparams- │ └── train.log- ├── ch_ppocr_server_v2.0_det_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ch_ppocr_server_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── chinese_cht_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── en_number_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── fa_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── french_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── german_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── hi_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── it_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── japan_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ka_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── korean_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── mr_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ne_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── oc_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── pu_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── rs_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── rsc_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ru_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ta_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── te_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ug_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── uk_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- ├── ur_mobile_v2.0_rec_train- │ ├── best_accuracy.pdopt- │ ├── best_accuracy.pdparams- │ ├── best_accuracy.states- │ └── train.log- └── xi_mobile_v2.0_rec_train- ├── best_accuracy.pdopt- ├── best_accuracy.pdparams- ├── best_accuracy.states- └── train.log- ```- </details>--* 把本项目`tools`文件夹里的`export_custom.py`复制到`PaddleOCR/tools`里,把`program.py`覆盖到`PaddleOCR/tools`里-* 把本项目`tools`文件夹里的`gen-models.sh`复制到`PaddleOCR`根目录-* 开始转换-- ```shell- cd PaddleOCR- chmod a+x gen-models.sh- ./gen-models.sh- ```--* 转换成功后,会在`PaddleOCR/inference`文件夹里生成31个onnx模型