diff --git a/README.md b/README.md index de674ed..05fbb8c 100755 --- a/README.md +++ b/README.md @@ -105,7 +105,7 @@ ## [在线demo](http://rapidocr.51pda.cn:9003/) - **说明**: 本在线demo不存储小伙伴们上传测试的任何图像数据 -- **demo所用模型组合为**: `ch_PP-OCRv3_det + ch_ppocr_mobile_v2.0_cls + ch_ppocr_mobile_v2.0_rec` +- **demo所用模型组合（最优组合）为**: `ch_PP-OCRv3_det + ch_ppocr_mobile_v2.0_cls + ch_PP-OCRv3_rec` - **运行机器配置**: `4核 AMD EPYC 7K62 48-Core Processor ` - **示例图**:
@@ -122,8 +122,8 @@ ├── assets # 一些演示用的图片，不是测试集 ├── commonlib # 通用库 ├── cpp # 基于c++的工程项目文件夹 + ├── docs # 相关的一些说明文档 ├── dotnet # .Net程序目录 - ├── FAQ.md # 一些问答整理 ├── images # 测试用图片，典型的测试图，一张是自然场景 ├── include # 编译c语言接口库时的头文件目录 ├── ios # 苹果手机平台工程目录 diff --git a/docs/README_en.md b/docs/README_en.md index 81db55f..e5e225b 100644 --- a/docs/README_en.md +++ b/docs/README_en.md @@ -1,6 +1,6 @@
- +
# RapidOCR @@ -8,7 +8,7 @@ [简体中文](../README.md) | English

- Open in Colab + Open in Colab @@ -33,7 +33,7 @@ - [Directory structure](#directory-structure) - [Current Progress](#current-progress) - [Model related](#model-related) - - [Download models( Google Drive|[Baidu NetDisk](https://pan.baidu.com/s/1mkirNltJS481In4g81jP3w?pwd=zy37) )](#download-models-google-drivebaidu-netdisk-) + - [Download models(Baidu NetDisk | [Google Drive](https://drive.google.com/drive/folders/1x_a9KpCo_1blxH1xFOfgKVkw1HYRVywY?usp=sharing))](#download-modelsbaidu-netdisk--google-drive) - [Model to onnx](#model-to-onnx) - [Compared](#compared) - [Text Det](#text-det) @@ -90,16 +90,16 @@ Since ubuntu users are all commercial users and have the ability to compile, pre | Platform | Compilation Status | Offer Status | | --------------- | -------- | -------- | -| Windows x86/x64 | [![CMake-windows-x86-x64](https://github.com/RapidOCR/RapidOCR/actions/workflows/windows-all-build.yaml/badge.svg)](https://github.com/RapidOCR/RapidOCR/actions/workflows/windows-all-build.yaml) | [Download Link](https://github.com/RapidAI/RapidOCR/releases) | -| Linux x64 | [![CMake-linux](https://github.com/RapidOCR/RapidOCR/actions/workflows/make-linux.yml/badge.svg)](https://github.com/RapidOCR/RapidOCR/actions/workflows/make-linux.yml) | Not available yet, compile by yourself | +| Windows x86/x64 | [![CMake-windows-x86-x64](https://github.com/RapidAI/RapidOCR/actions/workflows/windows-all-build.yaml/badge.svg)](https://github.com/RapidAI/RapidOCR/actions/workflows/windows-all-build.yaml) | [Download Link](https://github.com/RapidAI/RapidOCR/releases) | +| Linux x64 | [![CMake-linux](https://github.com/RapidAI/RapidOCR/actions/workflows/make-linux.yml/badge.svg)](https://github.com/RapidAI/RapidOCR/actions/workflows/make-linux.yml) | Not available yet, compile by yourself | ## Online demo - [Web demo](http://rapidocr.51pda.cn:9003/) -- The model combination used in the demo is: `ch_PP-OCRv3_det + ch_ppocr_mobile_v2.0_cls + ch_ppocr_mobile_v2.0_rec` +- The model combination (optimal combination) used for the demo is: `ch_PP-OCRv3_det + ch_ppocr_mobile_v2.0_cls + ch_PP-OCRv3_rec` - Environment: `4 Core AMD EPYC 7K62 48-Core Processor ` - **Gif Demo**:

- +
## Directory structure @@ -112,9 +112,8 @@ Since ubuntu users are all commercial users and have the ability to compile, pre ├── assets # Some pictures for demonstration, not a test set ├── commonlib # common library ├── cpp # C++-based project folder - ├── datasets # Additional training datasets + ├── docs # some docs ├── dotnet # .Net program directory - ├── FAQ.md # Some questions and answers ├── images # Test pictures, two typical test pictures, one is a natural scene, the other is a long text ├── include # The header file directory when compiling the c language interface library ├── ios # Apple mobile phone platform project directory @@ -138,12 +137,12 @@ Since ubuntu users are all commercial users and have the ability to compile, pre - [ ] 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 related -#### Download models( [Google Drive](https://drive.google.com/drive/folders/1x_a9KpCo_1blxH1xFOfgKVkw1HYRVywY?usp=sharing)|[Baidu NetDisk](https://pan.baidu.com/s/1mkirNltJS481In4g81jP3w?pwd=zy37) ) +#### Download models([Baidu NetDisk](https://pan.baidu.com/s/1mkirNltJS481In4g81jP3w?pwd=zy37) | [Google Drive](https://drive.google.com/drive/folders/1x_a9KpCo_1blxH1xFOfgKVkw1HYRVywY?usp=sharing)) #### Model to onnx - - ⭐[PaddleOCRModelConverter](https://github.com/RapidAI/PaddleOCRModelConverter) by @[SWHL](https://github.com/SWHL) + - [PaddleOCRModelConverter](https://github.com/RapidAI/PaddleOCRModelConverter) by @[SWHL](https://github.com/SWHL) - [Paddle2OnnxConvertor](https://github.com/RapidAI/Paddle2OnnxConvertor) by @[benjaminwan](https://github.com/benjaminwan) - - [Teach you to use ONNXRunTime to deploy PP-OCR](https://aistudio.baidu.com/aistudio/projectdetail/1479970?channelType=0&channel=0) by @Channingss + - [Teach you to use ONNXRunTime to deploy PP-OCR](https://aistudio.baidu.com/aistudio/projectdetail/1479970?channelType=0&channel=0) by @[Channingss](https://github.com/Channingss) ### Compared #### Text Det @@ -190,21 +189,21 @@ Since ubuntu users are all commercial users and have the ability to compile, pre - Scan the following QR code with QQ:
- +
## Demo #### Demonstration with C++/JVM
- +
#### Demonstration with .Net
- +
#### Demonstratioin with multi_language
- +
diff --git a/ocrweb/README.md b/ocrweb/README.md index 62b841e..ea76849 100755 --- a/ocrweb/README.md +++ b/ocrweb/README.md @@ -2,7 +2,7 @@ ### 相关问题 1. 各个阶段使用的模型以及配置参数有哪些？ - - 使用模型搭配为：`ch_PP-OCRv3_det + ch_ppocr_mobile_v2.0_cls + ch_ppocr_mobile_v2.0_rec` + - 使用模型搭配（最优组合）为：`ch_PP-OCRv3_det + ch_ppocr_mobile_v2.0_cls + ch_PP-OCRv3_rec` - 所有相关参数配置参见当前目录下的`config.yaml`文件 - 其中给出了使用模型，以及具体参数，参数具体介绍参见：[Link](https://github.com/RapidAI/RapidOCR/blob/main/python/README.md#configyaml%E4%B8%AD%E5%B8%B8%E7%94%A8%E5%8F%82%E6%95%B0%E4%BB%8B%E7%BB%8D) 2. 网页上显示的推理时间可以具体解释一下吗？ @@ -30,13 +30,12 @@ │   ├── ch_ppocr_v2_cls │   ├── ch_ppocr_v2_det │   ├── ch_ppocr_v2_rec - │   ├── ch_ppocr_v3_rec │   └── rapid_ocr_api.py ├── resources │   ├── models │ │   ├── ch_PP-OCRv3_det_infer.onnx │ │   ├── ch_ppocr_mobile_v2.0_cls_infer.onnx - │ │   └── ch_ppocr_mobile_v2.0_rec_infer.onnx + │ │   └── ch_PP-OCRv3_rec_infer.onnx │   └── rec_dict │ ├── en_dict.txt │ ├── japan_dict.txt diff --git a/python/README.md b/python/README.md index 8a052be..e66c5c2 100755 --- a/python/README.md +++ b/python/README.md @@ -18,6 +18,7 @@ ### 简介和说明 - **各个版本的ONNX模型下载地址：**[百度网盘](https://pan.baidu.com/s/1PTcgXG2zEgQU6A_A3kGJ3Q?pwd=jhai) | [Google Drive](https://drive.google.com/drive/folders/1x_a9KpCo_1blxH1xFOfgKVkw1HYRVywY?usp=sharing) - 所有常用的参数配置都在[`config.yaml`](./config.yaml)下，一目了然，更加便捷 +- **目前[`config.yaml`](./config.yaml)中配置为权衡速度和准确度的最优组合。** - 每个独立的模块下均有独立的`config.yaml`配置文件，可以单独使用 - `det`部分： - `det`中`mobile`和`server`版，推理代码一致，直接更改配置文件中模型路径即可 @@ -72,7 +73,6 @@ │   ├── ch_ppocr_v2_cls │   ├── ch_ppocr_v2_det │   ├── ch_ppocr_v2_rec - │   ├── ch_ppocr_v3_rec │   └── rapid_ocr_api.py ├── rapidocr_openvino │   ├── __init__.py @@ -88,15 +88,15 @@ │ ├── models │ │   ├── ch_PP-OCRv3_det_infer.onnx │ │   ├── ch_ppocr_mobile_v2.0_cls_infer.onnx - │ │   └── ch_ppocr_mobile_v2.0_rec_infer.onnx + │ │   └── ch_PP-OCRv3_rec_infer.onnx │ └── rec_dict │ ├── en_dict.txt │ ├── japan_dict.txt │ ├── korean_dict.txt │ └── ppocr_keys_v1.txt └── test_images - ├── det_images - └── rec_images + ├── ch_en_num.jpg + └── single_line_text.jpg ``` 3. 安装运行环境 - 基于onnxruntime推理所需环境安装： @@ -179,7 +179,7 @@ - `Rec`部分 | 参数名称 | 取值范围 | 默认值 | 作用 | | ------------: | :----------: | :-----: | :----------------------------------------------| - |`rec_img_shape`| - |`[3, 32, 320]`| 输入文本识别模型的图像Shape（CHW） | + |`rec_img_shape`| - |`[3, 48, 320]`| 输入文本识别模型的图像Shape（CHW） | |`rec_batch_num`| - | 6 | 批次推理的batch大小，一般采用默认值即可，太大并没有明显提速，效果还可能会差 | |`keys_path`| - | - | 文本识别模型推理所使用字典文件，始识别哪种类型文本而定（中英、日文等） |