Repository Issue Activity (beta)

#738 【功能增强建议】希望后续版本给engine()再加一个图片类型，就是传内存地址（和内存地址长度）

#739 Licence and notices for the three ONNX models bundled with rapidocr 3.9.2

#720 Arabic recognition cannot be paired with the multilingual detector — "Unsupported configuration"

#737 Make the OpenCV build selectable instead of hard-requiring opencv-python

#729 rapidocr v3.9.2 中 模型支持的lang_type有些混乱

#733 官方站点英文支持进行时

#723 Rapidocr with PP-OCRv6 small "rec" model less accurate than PP-OCRv5 "mobile" model with latin languages (missing interword spaces)

#724 延迟加载的并发安全

#726 返回单字框，英文单词背切分到字母细粒度

#714 Bug: TextClassifier ONNX session is eagerly loaded even when Global.use_cls is set to False

#719 Arabic recognition crashes: python-bidi is required at runtime but is not a declared dependency

#418 支持滑动窗口来识别较大图像

#419 rapidocr到2.0.7版本时，用TableStructureRec会出现错误

#537 有针对性初始化所需模型

#705 RapidOCR 默认参数与 PaddleOCR 参数对比

#708 是否单独对竖排文字做配置？

#695 TensorRT: multi_PP-OCRv3_det_mobile square-pad workaround can be eliminated by naming input H/W dimensions

#709 prepare_wheel_assets.py requires cv2 due to transitive import through __init__.py

#704 集成Paddle、PyTorch、MNN和TensorRT

#706 Bug: 略微倾斜的竖排文本框会导致检测框顶点顺序错误和识别结果崩坏

#701 PP-OCRv6模型在Rapidocr与paddle下识别效果不一致问题

#700 "日"字可能不会被识别

#686 support pp_ocr_v6

#697 Bug: latin cannot be resolved with version 3.9.0 of rapidocr

#694 'TRTInferSession' object has no attribute 'model_root_dir'