CVDec 16, 2023

ElasticLaneNet: An Efficient Geometry-Flexible Approach for Lane Detection

arXiv:2312.10389v22 citationsh-index: 9
Originality Incremental advance
AI Analysis

This addresses the problem of real-time lane detection for autonomous driving systems, offering a flexible method that is incremental in improving performance in complex scenarios.

The paper tackles lane detection with variable and complex geometries by proposing ElasticLaneNet, which uses an elastic interaction energy-loss function and achieves state-of-the-art results, including an F1-score of 89.51 on the SDLane dataset.

The task of lane detection involves identifying the boundaries of driving areas in real-time. Recognizing lanes with variable and complex geometric structures remains a challenge. In this paper, we explore a novel and flexible way of implicit lanes representation named \textit{Elastic Lane map (ELM)}, and introduce an efficient physics-informed end-to-end lane detection framework, namely, ElasticLaneNet (Elastic interaction energy-informed Lane detection Network). The approach considers predicted lanes as moving zero-contours on the flexibly shaped \textit{ELM} that are attracted to the ground truth guided by an elastic interaction energy-loss function (EIE loss). Our framework well integrates the global information and low-level features. The method performs well in complex lane scenarios, including those with large curvature, weak geometry features at intersections, complicated cross lanes, Y-shapes lanes, dense lanes, etc. We apply our approach on three datasets: SDLane, CULane, and TuSimple. The results demonstrate exceptional performance of our method, with the state-of-the-art results on the structurally diverse SDLane, achieving F1-score of 89.51, Recall rate of 87.50, and Precision of 91.61 with fast inference speed.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes