PaddleOCR icon indicating copy to clipboard operation
PaddleOCR copied to clipboard

Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and de...

Results 1088 PaddleOCR issues
Sort by recently updated
recently updated
newest added

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem - 系统环境/System Environment: - 版本号/Version:Paddle: PaddleOCR: 问题相关组件/Related components: - 运行指令/Command Code: - 完整报错/Complete Error Message: ![image](https://github.com/PaddlePaddle/PaddleOCR/assets/41010669/b342bdc2-829d-4126-9adb-027f67ba437f)

``` (paddle_env) ☁ PaddleOCR [main] ⚡ python tools/infer_kie_token_ser_re.py \ -o Architecture.Backbone.checkpoints=/Users/qyongkang/working/paddle/re_LayoutXLM_xfun_zh/ \ Global.infer_img=/Users/qyongkang/working/paddle/form/ \ -o_ser Architecture.Backbone.checkpoints=/Users/qyongkang/working/paddle/ser_LayoutXLM_xfun_zh/ Traceback (most recent call last): File "/Users/qyongkang/working/PaddleOCR/tools/infer_kie_token_ser_re.py", line 31, in import paddle ModuleNotFoundError: No...

Code PR is needed

微调det_v4,loss从降低4.9降到1.2,但是precision,recall,hmean都很低,基本是0.04,这种情况原因是啥?

用的官方默认模型,使用GPU时检测不到文字但CPU可以(无论官方样例图片还是自己图片都是这样)。 - 系统环境/System Environment:Windows 10 - 版本号/Version:PaddlePaddle-GPU:2.5.2 PaddleOCR:2.7.0.2 CPU下(`paddleocr --image_dir .\pages\01.jpg --use_gpu false`)输出: ``` [2024/05/14 08:17:25] ppocr DEBUG: dt_boxes num : 80, elapse : 1.6079998016357422 [2024/05/14 08:18:30] ppocr DEBUG: rec_res...

无法获取在线体验版的使用的模型版本,分别尝试了det模型的ch_PP-OCRv4 infer版和server_infer版,以及rec模型的infer版和server_infer版,发现效果均差于线上体验版。 例如识别图片: ![香菇粉](https://github.com/PaddlePaddle/PaddleOCR/assets/153589998/5051ea6e-440a-4586-8bb6-06e4ff5322ce) - 线上版结果: Black Sesame Mushroom Powder 黑芝麻香菇粉 ESTLADY -离线版结果: ('Black', 0.9862731099128723) ('Desame', 0.8805373311042786) ('Mushroom Powder', 0.9594960808753967) ('黑芝麻香菇粉', 0.9991157054901123) 对比结果离线版错将Sesame识别为Desame,漏掉了NESTLADY

paddleocr模型训练中用了预训练模型,但训练前期的acc为零,是否正常

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem - 系统环境/System Environment:Linux - 版本号/Version:Paddle: PaddleOCR: 问题相关组件/Related components: - 运行指令/Command Code: - 完整报错/Complete Error Message: 使用hubserving部署structure_table接口,用的模型是`ch_ppstructure_mobile_v2.0_SLANet`,调用后发现对于同一张表格图片识别效果一般,但是通用[表格识别demo](https://aistudio.baidu.com/community/app/91661/webUI)能识别正确,这是因为啥呢?

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem - 系统环境/System Environment: - 版本号/Version:Paddle: PaddleOCR: 问题相关组件/Related components: - 运行指令/Command Code: - 完整报错/Complete Error Message: 1.环境: win10、gpu、cuda11.2、cudnn8.2、paddocr2.7 2.数据: 按照ppocrlabel进行标注,并且显示没有问题; 3.训练:...

请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem - 系统环境/System Environment: - 版本号/Version:Paddle: PaddleOCR:2.7.5 问题相关组件/Related components: - 运行指令/Command Code: - 完整报错/Complete Error Message: - 字符间距大跟空格很像,造成空白区域被误识别成错误字符,如何解决?如何判断字符间距是否大,从而通过压扁图片,缩小间距改善结果? 原图 ![image](https://github.com/PaddlePaddle/PaddleOCR/assets/41010669/9e6e8751-009c-4e2b-b8fb-597b54edf48e) 识别结果有时会出现多余错误字符...

In case of ` det_box_type='poly'`, for some images, `np.array(offset.Execute(distance))` can result in inhomogeneous part of the detection box list, which cannot be casted into numpy array directly. Due to this...

contributor