paddleocr 3.4.0

Awesome multilingual OCR and document parsing toolkits based on PaddlePaddle

Stars: 70887, Watchers: 70887, Forks: 9836, Open Issues: 292

The PaddlePaddle/PaddleOCR repo was created 5 years ago and the last code push was Yesterday.
The project is extremely popular with a mindblowing 70887 github stars!

How to Install paddleocr

You can install paddleocr using pip

pip install paddleocr

or add it to a project with poetry

poetry add paddleocr

Package Details

Author

None

License

Apache License 2.0

Homepage

None

PyPi:

https://pypi.org/project/paddleocr/

GitHub Repo:

https://github.com/PaddlePaddle/PaddleOCR

Classifiers

• Utilities

Related Packages

Errors

A list of common paddleocr errors.

Code Examples

Here are some paddleocr code examples and snippets.

GitHub Issues

The paddleocr package has 292 open issues on GitHub

• [Feat] PaddleOCR-VL HPS Deployment
• fix docs
• 同一个图片,使用PPStructureV3识别,生成的json文件中会会出现formula_res_list,而使用PaddleOCRVL识别却不会
• Getting this error when picking a image
• 模型导出后使用和直接运行tools/infer_rec.py结果不同
• always Run Paddle-TRT Dynamic Shape mode.！！
• 一些相似文本会触发模型重复输出
• windows 上编译cpp_infer 文档不对,无法正常generate 生成vs solution(至少我电脑上一步步安装不对)
• 識別出來的效果特別差
• Segmentation fault in Docker container
• 昇腾910B服务器上PP-OCRv5_server_rec训练报错
• paddleocr-VL昇腾镜像的架构问题
• paddleocr ocr命令行方式进行文字识别和调用python3 tools/infer/predict_system.py 进行文字识别CPU使用率为什么不一样
• 按官方文档调用paddleocr-vl vllm出现连接未正常关闭导致服务不可用
• PPStructureV3 raise ValueError("The list of bounding boxes is empty.") when use_doc_unwarping=True for an image

See more issues on GitHub