An implementation on "I. Bogoslavskyi and C. Stachniss, "Fast range image-based segmentation of sparse 3D laser scans for online operation," 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Daejeon, 2016, pp. 163-169, doi: 10.1109/IROS.2016.7759050."
An official implementaiton is here:https://github.com/PRBonn/depth_clustering
The ground remove code is from https://github.com/AbangLZU/plane_fit_ground_filter. We achieve multi-plane fitting by using multi-thread.
1. We changed the ground segmentation method
2. Using hash map replace the image-based search
To the horizon neighbor search, we didn't use the method in paper. The method that we used to classify two neighbor objects as follows:
mkdir build
cd build
cmake ..
make
./range_cluster test.bin
