Starstruk Logo MarkStarstruk

v2.5.0

PaddlePaddle/PaddleOCRv2.5.0May 9, 2022by MissPenguin

AI Summary

This release focuses on performance improvements with the introduction of PP-OCRv3, which offers significant accuracy gains in Chinese and English scenes across 80 languages. Additionally, PPOCRLabelv2 is released to support table recognition and key information extraction annotation tasks.

Key Highlights

  • Release of PP-OCRv3 with improved accuracy in Chinese (+5%) and English (+11%) scenes.
  • Support for 80 languages with an average accuracy increase of over 5%.
  • Release of PPOCRLabelv2 for annotating table recognition, KIE, and irregular text.
  • Publication of the interactive e-book 'Dive into OCR'.

New Features

  • PP-OCRv3 models
  • PPOCRLabelv2 annotation tools
  • Annotation support for table recognition
  • Annotation support for key information extraction
  • Interactive e-book on OCR technology

Full Release Notes

**Release Note**

- Release [PP-OCRv3](./doc/doc_en/ppocr_introduction_en.md#pp-ocrv3): With comparable speed, the effect of Chinese scene is further improved by 5% compared with PP-OCRv2, the effect of English scene is improved by 11%, and the average recognition accuracy of 80 language multilingual models is improved by more than 5%.
- Release [PPOCRLabelv2](./PPOCRLabel): Add the annotation function for table recognition task, key information extraction task and irregular text image.
- Release interactive e-book [*"Dive into OCR"*](./doc/doc_en/ocr_book_en.md), covers the cutting-edge theory and code practice of OCR full stack technology.