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

Source code: https://github.com/xinge008/Cylinder3D

References in corpus (2)

• Tensor Low-Rank Reconstruction for Semantic Segmentation
• SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds

Cited by in corpus (12)

• (AF)2-S3Net: Attentive Feature Fusion with Adaptive Feature Selection for Sparse Semantic Segmentation Network
• AMVNet: Assertion-based Multi-View Fusion Network for LiDAR Semantic Segmentation
• LiDAR-based Panoptic Segmentation via Dynamic Shifting Network
• FusionPainting: Multimodal Fusion with Adaptive Attention for 3D Object Detection
• DRINet: A Dual-Representation Iterative Learning Network for Point Cloud Segmentation
• A Technical Survey and Evaluation of Traditional Point Cloud Clustering Methods for LiDAR Panoptic Segmentation
• SA-LOAM: Semantic-aided LiDAR SLAM with Loop Closure
• LiDAR R-CNN: An Efficient and Universal 3D Object Detector
• Input-Output Balanced Framework for Long-tailed LiDAR Semantic Segmentation
• Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perception
• S3Net: 3D LiDAR Sparse Semantic Segmentation Network
• Lite-HDSeg: LiDAR Semantic Segmentation Using Lite Harmonic Dense Convolutions