search

utiasSTARS / pykitti

Python tools for working with KITTI data.
MIT License
1.14k stars 239 forks source link

Creation of T_camX_velo all use R_rect_00 #34

Closed keenan-burnett closed 5 years ago

keenan-burnett commented 5 years ago

This is what is instructed in the paper, but this seems like a typo? I'm projecting the pointcloud onto the image plane for T_cam2_velo and I'm get slightly better results when using R_rect_20.

leeclemnet commented 5 years ago

I always thought this was strange, but the MATLAB devkit also uses R_rect_00 for all the velodyne-to-camera transformations. What do you mean when you say you get better results? Do you also get improvements for cam1 and cam3?

keenan-burnett commented 5 years ago

Which MATLAB devkit are you referring to? It seems that the velodyne points, when projected onto camera 2, align slightly better when R_rect_20 is used. I looked at the other cameras as well. Similarly for camera 3, the point projection works better when using R_rect_30. Surprisingly, using P_rect_30, with T_cam2_velo and R_rect_20 resulted in the best projection for camera 3.

projection_onto_cam2_with_r_rect_00 (P_rect_20) projection_onto_cam2_with_r_rect_00

projection_onto_cam2_with_r_rect_20 (P_rect_20) projection_onto_cam2_with_r_rect_20

projection_onto_cam3_with_t3_r_rect_00 (P_rect_30) projection_onto_cam3_with_t3_r_rect_00

projection_onto_cam3_with_t3_r_rect_30 (P_rect_30) projection_onto_cam3_with_t3_r_rect_30

projection_onto_cam3_with_t2_r_rect_20 (P_rect_30) projection_onto_cam3_with_t2_r_rect_20

leeclemnet commented 5 years ago

I'm referring to the raw data devkit provided on the KITTI website. If you look at the file loadCalibration.m, they've used R_rect_00 everywhere.

Looking at the KITTI dataset IJRR paper again (Equations 5-7), this seems to be because the way they build up the velodyne to camera transformation is with the sequence velodyne-->unrectified cam0 coordinates-->rectified cam0 coordinates-->rectified camN coordinates. The transformation from unrectified cam0 coordinates-->rectified cam0 coordinates should always use R_rect_00.

I think the improvements you're seeing might be coincidental, or else an artifact of imperfect calibration.

keenan-burnett commented 5 years ago

Okay, sounds good. Feel free to close.