GLane3D : Detecting Lanes with Graph of 3D Keypoints
Authors:
- 10 June 2025
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 27508-27518
Abstract
Accurate and efficient lane detection in 3D space is essential for autonomous driving systems, where robust generalization is the foremost requirement for 3D lane detection algorithms. Considering the extensive variation in lane structures worldwide, achieving high generalization capacity is particularly challenging, as algorithms must accurately identify a wide variety of lane patterns worldwide. Traditional top-down approaches rely heavily on learning lane characteristics from training datasets, often struggling with lanes exhibiting previously unseen attributes. To address this generalization limitation, we propose a method that detects keypoints of lanes and subsequently predicts sequential connections between them to construct complete 3D lanes. Each key point is essential for maintaining lane continuity, and we predict multiple proposals per keypoint by allowing adjacent grids to predict the same keypoint using an offset mechanism. PointNMS is employed to eliminate overlapping proposal keypoints, reducing redundancy in the estimated BEV graph and minimizing computational overhead from connection estimations. Our model surpasses previous state-of-the-art methods on both the Apollo and OpenLane datasets, demonstrating superior F1 scores and a strong generalization capacity when models trained on OpenLane are evaluated on the Apollo dataset, compared to prior approaches.Keywords
All Related Versions
This publication has 44 references indexed in Scilit:
- PETR: Position Embedding Transformation for Multi-view 3D Object Detection2022
- PersFormer: 3D Lane Detection via Perspective Transformer and the OpenLane Benchmark2022
- RCLane: Relay Chain Prediction for Lane Detection2022
- Polynomial Regression Network for Variable-Number Lane Detection2020
- Ultra Fast Structure-Aware Deep Lane Detection2020
- CurveLane-NAS: Unifying Lane-Sensitive Architecture Search and Adaptive Point Blending2020
- Gen-LaneNet: A Generalized and Scalable Approach for 3D Lane Detection2020
- Lane Detection and Classification Using Cascaded CNNs2020
- EL-GAN: Embedding Loss Driven Generative Adversarial Networks for Lane Detection2019
- Inverse perspective mapping simplifies optical flow computation and obstacle detectionBiological Cybernetics, 1991