Laserscanner: Velodyne HDL-64E
1、垂直视角为26.9度，分辨率为0.4度。垂直视野被分成传感器上方+2度和传感器下方-24.9度。
2、360度的水平视野，分辨率为0.08-0.35（取决于旋转速度）
3、旋转速度可以选择在5-20Hz之间。
The laser scanner spins at 10 frames per second, capturing approximately 100k points per cycle. The vertical resolution of the laser scanner is 64. The cameras are mounted approximately level with the ground plane. The camera images are cropped to a size of 1382 x 512 pixels
Some papers’ choice: 0-16 for training, 17-18 for validation and 19-20 for testing
Point distribution of KITTI cars of 00-20 sequences
7481 training images and 7518 test images as well as the corresponding point clouds, comprising a total of 80.256 labeled objects.
KITTI dataset [7] only provides annotations in front camera’s field of view, its detection range is set to be [0, 70.4m] for the X axis, [-40m, 40m] for the Y axis, and [-3m, 1m] for the Z axis. We set the voxel size to be (0.05m, 0.05m, 0.1m) =》1400160040
21 training sequences and 29 test sequences.
labeled 8 different classes, only the classes 'Car' and 'Pedestrian' are evaluated in our benchmark,