GitLab

fix(batch): disable OCR detection batch processing by default

支持batch-ocr-det，速度约提升3倍（200页pdf在3090上）

master->dev

Release 1.3.12

feat(docs): update changelog for PP-OCRv5 model support and handwritten document recognition enhancements

feat(docs): update changelog for PP-OCRv5 model support and handwritten document recognition enhancements

fix(ocr): adjust area ratio threshold and update fitz document handling in image conversion

Merge pull request #10 from myhloli/img2text

feat(ocr): add area ratio calculation for OCR results and enhance get_coords_and_area function

feat(ocr): add area ratio calculation for OCR results and enhance get_coords_and_area function

feat(ocr): implement PPHGNetV2 architecture with multiple stages and layers

feat(ocr): add PPHGNetV2_B4 backbone and update OCR models

- Add PPHGNetV2_B4 backbone to the list of supported backbones
- Introduce new OCR model configuration for PP-OCRv5 with PPHGNetV2_B4
- Update existing model configurations to use the new backbone
- Modify RNN neck to support input with H > 1
- Adjust batch size for inference

feat(ocr): add PP-OCRv5 models and update configurations

- Add new PP-OCRv5 detection and recognition models
- Update arch_config.yaml with new model architectures
- Modify models_config.yml to include PP-OCRv5 models for ch_lite configuration- Change dictionary file for ch_lite to ppocrv5_dict.txt

fix(ocr_mkcontent): improve image handling and footnote integration in markdown output

master->dev

Release 1.3.11

docs(changelog): remove pdfminer.six version pinning from release notes

docs(changelog): remove pdfminer.six version pinning from release notes

Release 1.3.11

fix(modeling): escape backslashes in LaTeX command descriptions

fix(requirements): update pdfminer.six version and restrict torch version upper limit

fix(setup): update python_requires to support Python 3.10 to 3.13

fix(Dockerfile): update modelscope installation command to use mirror