Hello,
I got several question to your dataset. It would be a big help, if you could answer them.
Did you used the same classes (Car, Pedestrian, Cyclist etc.) like in the KITTI dataset, expect of some exceptions?
How are the classes distributed (class balance) in the whole dataset?
Can you provide a range like in KITTI? For example KITTI paper: x(0, 69.12), y(-39.68, 39.68), z(-3, 1)
Thanks in advance