GitLab

refactor: enhance markdown generation by introducing pipeline_union_make and improving latex delimiter handling

refactor: improve image handling by transitioning from NumPy arrays to PIL images in cropping functions

refactor: enhance image margin cropping and processing for improved handling of PIL and NumPy images

refactor: enhance document parsing by supporting multiple PDF files and improving method organization

feat: add mineru-vlm backend.

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