Cylinder3D: An Effective 3D Framework for Driving-scene LiDAR Semantic Segmentation

TL;DR

Cylinder3D introduces a novel 3D LiDAR segmentation framework that preserves 3D topology, outperforming 2D projection methods and achieving state-of-the-art results on SemanticKITTI.

Contribution

The paper develops a 3D cylinder-based representation and convolution framework, along with a dimension-decomposition context module, to enhance LiDAR segmentation accuracy.

Findings

Outperforms existing methods by 6% mIoU on SemanticKITTI

Effectively exploits 3D topology in point cloud segmentation

Achieves state-of-the-art performance in driving-scene LiDAR segmentation

Abstract

State-of-the-art methods for large-scale driving-scene LiDAR semantic segmentation often project and process the point clouds in the 2D space. The projection methods includes spherical projection, bird-eye view projection, etc. Although this process makes the point cloud suitable for the 2D CNN-based networks, it inevitably alters and abandons the 3D topology and geometric relations. A straightforward solution to tackle the issue of 3D-to-2D projection is to keep the 3D representation and process the points in the 3D space. In this work, we first perform an in-depth analysis for different representations and backbones in 2D and 3D spaces, and reveal the effectiveness of 3D representations and networks on LiDAR segmentation. Then, we develop a 3D cylinder partition and a 3D cylinder convolution based framework, termed as Cylinder3D, which exploits the 3D topology relations and structures of driving-scene point clouds. Moreover, a dimension-decomposition based context modeling module is introduced to explore the high-rank context information in point clouds in a progressive manner. We evaluate the proposed model on a large-scale driving-scene dataset, i.e. SematicKITTI. Our method achieves state-of-the-art performance and outperforms existing methods by 6% in terms of mIoU.