Advancements in 3D Lane Detection Using LiDAR Point Clouds: From Data Collection to Model Development

TL;DR

This paper introduces LiSV-3DLane, a comprehensive 360-degree LiDAR dataset for 3D lane detection, and proposes LiLaDet, a novel model that leverages LiDAR geometry for improved accuracy in complex driving environments.

Contribution

The paper presents a large-scale, full-surround LiDAR dataset with semantic annotations and a new 3D lane detection model utilizing LiDAR geometry, advancing 3D lane detection research.

Findings

LiLaDet outperforms existing methods on K-Lane and LiSV-3DLane datasets.

LiSV-3DLane provides a full 360-degree view with complex lane patterns.

Automatic annotation pipeline generates finer lane labels efficiently.

Abstract

Advanced Driver-Assistance Systems (ADAS) have successfully integrated learning-based techniques into vehicle perception and decision-making. However, their application in 3D lane detection for effective driving environment perception is hindered by the lack of comprehensive LiDAR datasets. The sparse nature of LiDAR point cloud data prevents an efficient manual annotation process. To solve this problem, we present LiSV-3DLane, a large-scale 3D lane dataset that comprises 20k frames of surround-view LiDAR point clouds with enriched semantic annotation. Unlike existing datasets confined to a frontal perspective, LiSV-3DLane provides a full 360-degree spatial panorama around the ego vehicle, capturing complex lane patterns in both urban and highway environments. We leverage the geometric traits of lane lines and the intrinsic spatial attributes of LiDAR data to design a simple yet effective automatic annotation pipeline for generating finer lane labels. To propel future research, we propose a novel LiDAR-based 3D lane detection model, LiLaDet, incorporating the spatial geometry learning of the LiDAR point cloud into Bird's Eye View (BEV) based lane identification. Experimental results indicate that LiLaDet outperforms existing camera- and LiDAR-based approaches in the 3D lane detection task on the K-Lane dataset and our LiSV-3DLane.