/ Machine Learning

Semantic and Instance Segmentation of LiDAR point clouds for autonomous driving

Semantic and Instance Segmentation of LiDAR point clouds for autonomous driving

LiDAR-Bonnetal

Semantic Segmentation of point clouds using range images.

Developed by Andres Milioto, Jens Behley, Ignacio Vizzo, and Cyrill Stachniss

Examples of segmentation results from SemanticKITTI dataset: ptcl ptcl

This code provides code to train and deploy Semantic Segmentation of LiDAR scans, using range images as intermediate representation. The training pipeline can be found in /train. We will open-source the deployment pipeline soon.

semantic-ptcl

GitHub