On Deep Learning for Geometric and Semantic Scene Understanding Using On-Vehicle 3D LiDAR

TL;DR

This paper introduces DurLAR, a high-fidelity 3D LiDAR dataset and a novel segmentation pipeline that enhances accuracy and efficiency in geometric and semantic scene understanding for autonomous driving.

Contribution

The paper presents DurLAR, the first high-fidelity 128-channel LiDAR dataset, and a new segmentation method using RAPiD features and a smaller architecture for improved accuracy and efficiency.

Findings

DurLAR dataset enables better scene understanding.

Proposed pipeline achieves higher segmentation accuracy with fewer annotations.

Method improves efficiency in 3D LiDAR segmentation tasks.

Abstract

3D LiDAR point cloud data is crucial for scene perception in computer vision, robotics, and autonomous driving. Geometric and semantic scene understanding, involving 3D point clouds, is essential for advancing autonomous driving technologies. However, significant challenges remain, particularly in improving the overall accuracy (e.g., segmentation accuracy, depth estimation accuracy, etc.) and efficiency of these systems. To address the challenge in terms of accuracy related to LiDAR-based tasks, we present DurLAR, the first high-fidelity 128-channel 3D LiDAR dataset featuring panoramic ambient (near infrared) and reflectivity imagery. To improve efficiency in 3D segmentation while ensuring the accuracy, we propose a novel pipeline that employs a smaller architecture, requiring fewer ground-truth annotations while achieving superior segmentation accuracy compared to contemporary approaches. To improve the segmentation accuracy, we introduce Range-Aware Pointwise Distance Distribution (RAPiD) features and the associated RAPiD-Seg architecture. All contributions have been accepted by peer-reviewed conferences, underscoring the advancements in both accuracy and efficiency in 3D LiDAR applications for autonomous driving. Full abstract: https://etheses.dur.ac.uk/15738/.