Deep Learning for LiDAR Point Clouds in Autonomous Driving: A Review

TL;DR

This paper systematically reviews deep learning methods applied to LiDAR point clouds in autonomous driving, covering architectures, tasks, datasets, and performance, highlighting recent advances and future challenges.

Contribution

It provides the first comprehensive survey of deep learning techniques for LiDAR point clouds in autonomous driving, summarizing 140 key contributions from recent years.

Findings

Deep learning architectures have significantly advanced LiDAR data processing.

State-of-the-art methods achieve high accuracy in segmentation and detection.

Remaining challenges include data noise and real-time processing requirements.

Abstract

Recently, the advancement of deep learning in discriminative feature learning from 3D LiDAR data has led to rapid development in the field of autonomous driving. However, automated processing uneven, unstructured, noisy, and massive 3D point clouds is a challenging and tedious task. In this paper, we provide a systematic review of existing compelling deep learning architectures applied in LiDAR point clouds, detailing for specific tasks in autonomous driving such as segmentation, detection, and classification. Although several published research papers focus on specific topics in computer vision for autonomous vehicles, to date, no general survey on deep learning applied in LiDAR point clouds for autonomous vehicles exists. Thus, the goal of this paper is to narrow the gap in this topic. More than 140 key contributions in the recent five years are summarized in this survey, including the milestone 3D deep architectures, the remarkable deep learning applications in 3D semantic segmentation, object detection, and classification; specific datasets, evaluation metrics, and the state of the art performance. Finally, we conclude the remaining challenges and future researches.