Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perception

TL;DR

This paper introduces a novel cylindrical and asymmetrical 3D convolution network for outdoor LiDAR perception, effectively capturing 3D geometric patterns while respecting the sparsity and density variations of outdoor point clouds.

Contribution

The paper proposes a new framework with cylindrical partitioning and asymmetrical 3D convolutions, achieving state-of-the-art results across multiple LiDAR perception tasks.

Findings

State-of-the-art on SemanticKITTI leaderboard

Significant improvements on nuScenes and A2D2 datasets

Strong generalization to panoptic segmentation and 3D detection

Abstract

State-of-the-art methods for driving-scene LiDAR-based perception (including point cloud semantic segmentation, panoptic segmentation and 3D detection, \etc) often project the point clouds to 2D space and then process them via 2D convolution. Although this cooperation shows the competitiveness in the point cloud, it inevitably alters and abandons the 3D topology and geometric relations. A natural remedy is to utilize the 3D voxelization and 3D convolution network. However, we found that in the outdoor point cloud, the improvement obtained in this way is quite limited. An important reason is the property of the outdoor point cloud, namely sparsity and varying density. Motivated by this investigation, we propose a new framework for the outdoor LiDAR segmentation, where cylindrical partition and asymmetrical 3D convolution networks are designed to explore the 3D geometric pattern while maintaining these inherent properties. The proposed model acts as a backbone and the learned features from this model can be used for downstream tasks such as point cloud semantic and panoptic segmentation or 3D detection. In this paper, we benchmark our model on these three tasks. For semantic segmentation, we evaluate the proposed model on several large-scale datasets, \ie, SemanticKITTI, nuScenes and A2D2. Our method achieves the state-of-the-art on the leaderboard of SemanticKITTI (both single-scan and multi-scan challenge), and significantly outperforms existing methods on nuScenes and A2D2 dataset. Furthermore, the proposed 3D framework also shows strong performance and good generalization on LiDAR panoptic segmentation and LiDAR 3D detection.