Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs

TL;DR

This paper introduces a deep learning framework using superpoint graphs to efficiently perform semantic segmentation on large-scale 3D point clouds, significantly improving accuracy over previous methods.

Contribution

The novel superpoint graph structure captures scene organization, enabling effective graph convolutional network-based segmentation of large point clouds.

Findings

Achieved state-of-the-art results on Semantic3D outdoor LiDAR data

Improved indoor scan segmentation accuracy on S3DIS dataset

Enhanced segmentation performance by capturing contextual relationships

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).