How to convert instance results to 3D bounding boxes of Object detection?

[shuluoshu]

Hi, @laughtervv , @Rotch ,
I wonder if you could offer the code for converting instance segmentation results into 3D bounding boxes of object detection, since I don't know how to compute the orientation of the bounding box.
Thanks a lot!

[CrazySnailer]

Drawing 3D bounding boxes need priori knowledge, Like Frustum Pointnet, the orientation of the bounding box need to be regression by the priori knowledge. Look forward to discuss with you.

[laughtervv]

In our detection experiment, we didn't consider the box orientation. Both ground truth and the predicted bounding boxes are computed by the min-max xyz value of point cloud instances.

[CrazySnailer]

@laughtervv Thanks.