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:windows - 版本号/Version:Paddle:2.6 PaddleOCR: 问题相关组件/Related components: - 运行指令/Command Code:python3 tools/train.py -c configs/det/det_r50_vd_dcn_fce_ctw.yml - 完整报错/Complete Error Message: Traceback (most...

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+gpu - 版本号/Version : Paddlehub:2.3.0 PaddleOCR:2.6.0.1 问题相关组件/Related components:hubserving - 运行指令/Command Code: 服务启动 `` hub serving start -c deploy/hubserving/structure_system/config.json...

标注的是16个点的坐标,文字内容不好切分出来构建数据集。db++,检测出来的弯曲文本框,改如何让svrt,进行识别。get_rotate_crop_image()这边也是裁剪的,4坐标点。

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:centos 7.9 - 版本号/Version:Paddle:2.3.1 PaddleOCR: 2.6.0.1 问题相关组件/Related components: - 运行指令/Command Code:paddleocr --image_dir=ppstructure/docs/table/1.png --type=structure - 完整报错/Complete Error Message: '==SUPPRESS==',...

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: 系统:win11 paddle: paddlepaddle-gpu 2.3.2 paddleOCR:...

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: 增加了90°和270°两个角度并对模型进行训练,在python上验证无误,可以输出90°和270°结果,但在cpp部署的官方样例程序上,该模型经过推理后的推理模型准确度却下降了,请问为什么方向分类器推理后在cpp上的准确度不如训练模型在python上的准确度呢? ![1666144706607](https://user-images.githubusercontent.com/108849683/196579439-ad17579c-7001-4743-a50b-640c98bf3322.png) ![1666144229641](https://user-images.githubusercontent.com/108849683/196579443-46165b1e-31b1-4f93-ba06-ffee827e71f2.png)

status/close

Please provide the following information to quickly locate the problem - System Environment:Amazon Linux 2 - Version:Paddle:PaddlePaddle 2.3.2 PaddleOCR: paddleocr==2.6.0.2 Command Code:CUDA_VISIBLE_DEVICES=0 python3 tools/infer/predict_system.py --image_dir="./doc/imgs_en/254.jpg" --det_model_dir="./en_PP-OCRv3_det_slim_infer/" --rec_model_dir="./en_PP-OCRv3_rec_slim_infer/" --rec_char_dict_path="./ppocr/utils/en_dict.txt" 完整报错/Complete Error...

同样的训练数据集,V2版本模型训练精度能收敛到0.89左右,而V3版本的模型精度只能收敛到0.80左右。 使用的是默认的ch_PP-OCRv3_rec_distillation.yml配置,由于显卡显存不够,只修改了batch size从128减小到64。 请问是什么原因了,需要做哪些修改? 怀疑是GTC训练策略有影响,现在只想训练不带GTC只用CTC的PP-OCRV3识别模型,修改配置文件会报错,看报错信息是训练V3必须带GTC的部分loss的内容。请问怎样修改配置才能训练不带GTC的识别模型,或者给出只用CTC训练的训练配置文件?

recognition