boxes_preds:
seed_points
seed_points
vote_points
vote_features
vote_offset
aggregated_points
aggregated_features
obj_scores
center_offset
center
size
dir_class
对于网络检测输出的bbox,具体的打印格式如下
限制了score>0.5
<class 'list'>
[{'boxes_3d': LiDARInstance3DBoxes(
tensor([[30.2626, -0.6454, -0.7021, 1.6633, 3.5731, 1.5774, 1.5827],
[36.6007, 3.8564, -0.6431, 1.6344, 3.8559, 1.5411, 4.6442],
[24.6624, 3.8071, -0.8209, 1.6143, 4.0631, 1.5424, 4.6454],
[17.8412, 3.8099, -0.9454, 1.6326, 4.1052, 1.4439, 4.7972]])), 'scores_3d': tensor([0.8437, 0.6146, 0.8867, 0.9141]), 'labels_3d': tensor([0, 0, 0, 0])}]
up z x front (yaw=-0.5*pi)
^ ^
| /
| /
(yaw=-pi) left y <------ 0 -------- (yaw=0)
Each row is (x, y, z, x_size, y_size, z_size, yaw)
yaw加了一个pi????
bbox的转换函数:
Box3DMode.convert
Coordinates in LiDAR:
.. code-block:: none
up z
^ x front
| /
| /
left y <------ 0
The relative coordinate of bottom center in a LiDAR box is (0.5, 0.5, 0),
and the yaw is around the z axis, thus the rotation axis=2.
Coordinates in camera:
.. code-block:: none
z front
/
/
0 ------> x right
|
|
v
down y
The relative coordinate of bottom center in a CAM box is [0.5, 1.0, 0.5],
and the yaw is around the y axis, thus the rotation axis=1.
Coordinates in Depth mode:
.. code-block:: none
顺时针为正
up z
^ y front
| /
| /
0 ------> x right
The relative coordinate of bottom center in a DEPTH box is (0.5, 0.5, 0),
and the yaw is around the z axis, thus the rotation axis=2.
"""
在mmdetection3d中,最后现实的坐标系都是在depth坐标系下,如果想输出和kitti数据集label相同坐标系的预测,可以在show_result中进行输出为.txt文件。
搞清楚