English

Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs

Computer Vision and Pattern Recognition 2018-03-29 v2 Machine Learning Neural and Evolutionary Computing

Abstract

We propose a novel deep learning-based framework to tackle the challenge of semantic segmentation of large-scale point clouds of millions of points. We argue that the organization of 3D point clouds can be efficiently captured by a structure called superpoint graph (SPG), derived from a partition of the scanned scene into geometrically homogeneous elements. SPGs offer a compact yet rich representation of contextual relationships between object parts, which is then exploited by a graph convolutional network. Our framework sets a new state of the art for segmenting outdoor LiDAR scans (+11.9 and +8.8 mIoU points for both Semantic3D test sets), as well as indoor scans (+12.4 mIoU points for the S3DIS dataset).

Keywords

Cite

@article{arxiv.1711.09869,
  title  = {Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs},
  author = {Loic Landrieu and Martin Simonovsky},
  journal= {arXiv preprint arXiv:1711.09869},
  year   = {2018}
}

Comments

Accepted to CVPR 2018; camera ready version. Major updates to [v1]: Improved performance on S3DIS (from +5.8 to +12.4 mIoU) and extended ablation study in Appendix

R2 v1 2026-06-22T22:58:20.468Z