chineseocr_lite icon indicating copy to clipboard operation
chineseocr_lite copied to clipboard

超轻量级中文ocr,支持竖排文字识别, 支持ncnn、mnn、tnn推理 ( dbnet(1.8M) + crnn(2.5M) + anglenet(378KB)) 总模型仅4.7M

Results 135 chineseocr_lite issues
Sort by recently updated
recently updated
newest added

运行后出现如下报错,不知道是什么原因 2021-11-17 20:10:41.0305976 [W:onnxruntime:, graph.cc:3391 onnxruntime::Graph::CleanUnusedInitializers] Removing initializer '88'. It is not used by any node and should be removed from the model. 2021-11-17 20:10:41.0306080 [W:onnxruntime:, graph.cc:3391 onnxruntime::Graph::CleanUnusedInitializers] Removing initializer...

[web](https://github.com/DayBreak-u/chineseocr_lite/blob/3f0a6c40873b36c29c84223860d7807a6da77c43/backend/webInterface/tr_run.py#L51) 服务里的参数实际为`img`,所以示例参数里的`image`应该为`img`。

![image](https://user-images.githubusercontent.com/57654726/76915596-be578f00-68b5-11ea-8405-f1a7d330fe45.png)

Can you guys help me train the model with my own data?

对于印刷体识别效果已经很好了,能否有个教程指导下怎么自己训练模型的,谢谢。

现在身份证识别效果咋样?

chineseocr_lite有没有模型只识别英文的呢?

Who can share the APK installation package of the OCR demo generated by TNN, I want to test it.

程序在开发电脑上运行正常,拷贝到其他电脑时运行报错,请大神指导下,谢谢 ![1634180708](https://user-images.githubusercontent.com/26628875/137245633-88e99d03-77d6-461e-87e1-169db4754e3d.png)

chineseocr 项目中没有提供这个轻量版模型,请问如何训练这个轻量版的模型?谢谢!