ddw

數據預處理

3

在train.py 裏直接加載的預處理後的數據，但預處理直接按最大邊等比縮放到600，對於不同長寬比的數據這樣處理不太合適吧，感覺padding應該更好一點

the project first use chessboard to calibration camera intrinsic, and then get pixel coordinate of corners from well calibrated images with OptimalNewCameraMatrix to calculate extrinsic in the code, but in...

怎么使用你的数据呢，图像image和机械臂pose是怎么对应的，对于精度是怎么验证的呢

I calibrate SensorX2car/lidar2car using the data you provided for example test, after running the code, I got wrong pose containing lots of 1 0 0 0 0 1 0 0...