v2.0.0
PaddlePaddle/PaddleOCRv2.0.0Feb 8, 2021by MissPenguin
AI Summary
PaddleOCR v2.0.0 is a major release that introduces dynamic graph programming paradigm support adapted to Paddle 2.0, featuring multiple detection (DB, EAST, SAST) and recognition (Rosetta, CRNN, RARE, SRN, STAR-Net) algorithms, along with PPOCR Chinese models and 25 multilingual models. The release also includes deployment adaptations for Inference (Python/C++), whl, and serving, plus new data annotation (PPOCRLabel) and synthesis (Style-Text) tools.
Key Highlights
- Dynamic graph programming paradigm support adapted to Paddle 2.0
- Multiple detection algorithms: DB, EAST, SAST
- Multiple recognition algorithms: Rosetta, CRNN, RARE, SRN, STAR-Net
- PPOCR Chinese models with mobile and server variants for detection, classification, and recognition
- Multilingual support for 25 languages including English, Japanese, Korean, French, and German
Breaking Changes
- Migration to Paddle 2.0 dynamic graph paradigm (may require code adjustments for users upgrading from v1.x)
New Features
- Dynamic graph programming paradigm support
- Detection algorithms: DB, EAST, SAST
- Recognition algorithms: Rosetta, CRNN, RARE, SRN, STAR-Net
- PPOCR Chinese detection models (mobile, server)
- PPOCR text direction classification models (mobile)
- PPOCR recognition models (mobile, server)
- Multilingual English model (mobile)
- Multilingual models for 25 languages (mobile)
- Deployment support: Inference (Python and C++), whl package, serving
- Style-Text data synthesis tool for generating target scene images
- PPOCRLabel data annotation tool with direct training compatibility
Full Release Notes
**Release Note**
一、Support dynamic graph programming paradigm, adapted to Paddle 2.0, including:
1. Detection algorithm: DB, EAST, SAST
2. Recognition algorithm: Rosetta, CRNN, RARE, SRN, STAR-Net
3. PPOCR Chinese models：
(1) Detection models: mobile, server
(2) Text direction classification models: mobile
(3) Recognition models: mobile, server
4. Multilingual models:
(1) English: mobile
(2) Japanese, Korean, French, German, etc. 25 languages in total: mobile
二、The related works on deployment have been well adapted, including Inference(Python, C++) , whl, and serving
三、Release the annotation and synthesis tools:
1. Release a new data synthesis tool, i.e., Style-Text，easy to synthesize a large number of images which are similar to the target scene image.
2. Release a new data annotation tool, i.e., PPOCRLabel, which is helpful to improve the labeling efficiency. Moreover, the labeling results can be used in training of the PP-OCR system directly.