LPRNet_Pytorch

LPRNet_Pytorch copied to clipboard

Enhancement: Expanding LPRNet's Recognition Capabilities to Global License Plate Varieties

1

Dear LPRNet Contributors, First and foremost, I would like to extend my sincerest commendations for developing such a robust and lightweight framework for license plate recognition. The performance metrics and...

yihong1120

请问一下数据集怎么标注

1

请问一下数据集怎么标注?需要用CCPD数据集吗

programcj

我的loss从2、3降到0.5左右，acc从0加到0.2，然后loss开始变成负数了，很奇怪！acc也从0.2突变成0

1

zhaojiankuan

https://github.com/pcb9382/PlateRecognition

pcb9382

some times loss gets very high. Is there any chances that train stops and outputs best model at the moment loss is very high? If so, it would be neccesary...

jeungmin717

数据集

1

可提供车牌黄绿蓝100W数据集q:1041357701

daijie1223

@sirius-ai Hello, Thanks for this wonderful repository In my case, KOR licenseplate has 7digits or 8digits. Type A , and Type B Type A is like 12A1234 Type B is...

jeungmin717

Fix opencv imread when the fpath contains Chinese

1

opencv 无法读取中文路径下的图片(如Windows系统上）。使用 numpy 读取到内存，然后使用 opencv 解码，实现图片读取。

ausk

Is there confidence in license plate recognition

ZhaoMonica

为什么训练和测试的时候，没有使用model.train()和model.eval()

petervxy