I am a Master student working on my Master thesis. My objective is to develop a target recognition framework for LiDAR data based on 3D Krawtchouk. I aim to isolate the point cloud of each bounding box so to train my classifier after Krawtchouk is applied. I can work in Python or MATLAB. Any suggestions?
Hi, for each pointcloud and each of its boxes, run https://github.com/nutonomy/nuscenes-devkit/blob/master/python-sdk/nuscenes/utils/geometry_utils.py#L111 and save the points that are inside the box.