search

tsinghua-rll / VoxelNet-tensorflow

A 3D object detection system for autonomous driving.
MIT License
453 stars 123 forks source link

camera to lidar coordinate convertion for another lidar sensor #43

Closed a-ngo closed 1 year ago

a-ngo commented 6 years ago

how did u come up with the matrices for camera and lidar coordinate convertion in config.py? (__C.MATRIX_P2, C.MATRIX_T_VELO_2_CAM, C.MATRIX_R_RECT_0 )

the thing is I want to try the VoxelNet with another lidar sensor but I don't know how to adjust this matrices.

xueyingliu commented 6 years ago

@artdengo I have the same question with you, have you already solved it?

LinHungShi commented 6 years ago

I think the author just calculates the average of each matrix over all images. The calibration matrices for each image and lidar can be downloaded from kitti website.

a-ngo commented 6 years ago

@xueyingliu I didn't solve it yet.

@LinHungShi so do you think that it is possible to completely remove the usage of these calibration matrices?? and how would you do that? because I got my own lidar and label data, that are already calibrated and I assume that images are only used for visualization purposes, no?

LinHungShi commented 6 years ago

if you want to detect objects in lidar, you don't need those matrices. calibration is needed because the label in kitti is in camera coordinate.

xueyingliu commented 6 years ago

@LinHungShi really thanks for your reply, now it is clear.