基于PaddleOCR解决文本检测训练模型与inference模型预测效果不一致的问题更简单的解决方法
PaddleOCR
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 deployment among server, mobile, embedded and IoT devices)
项目地址:https://gitcode.com/gh_mirrors/pa/PaddleOCR

·
解决方法
在模型预测时加入参数–det_limit_type=“resize_long”
PaddleOCR
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 deployment among server, mobile, embedded and IoT devices)
项目地址:https://gitcode.com/gh_mirrors/pa/PaddleOCR
这样在不需要修改代码也可以解决该问题.




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 deployment among server, mobile, embedded and IoT devices)
最近提交(Master分支:8 个月前 )
934faf59
3 天前
332d9d51
* docs: Fix formatting
* Fix typo
* Fix translation
* Fix formatting
* Fix formatting 14 天前
更多推荐
目录
所有评论(0)