LDTR:Transformer-based lane detection with anchor-chain representation  

作　　者：Zhongyu Yang Chen Shen Wei Shao Tengfei Xing Runbo Hu Pengfei Xu Hua Chai Ruini Xue 

机构地区：[1]School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China [2]Didi Chuxing,Beijing 100081,China

出　　处：《Computational Visual Media》2024年第4期753-769,共17页计算可视媒体（英文版）

基　　金：supported by the National Natural Science Foundation of China(No.U23A6007).

摘　　要：Despite recent advances in lane detection methods,scenarios with limited-or no-visual-clue of lanes due to factors such as lighting conditions and occlusion remain challenging and crucial for automated driving.Moreover,current lane representations require complex post-processing and struggle with specific instances.Inspired by the DETR architecture,we propose LDTR,a transformer-based model to address these issues.Lanes are modeled with a novel anchorchain,regarding a lane as a whole from the beginning,which enables LDTR to handle special lanes inherently.To enhance lane instance perception,LDTR incorporates a novel multi-referenced deformable attention module to distribute attention around the object.Additionally,LDTR incorporates two line IoU algorithms to improve convergence efficiency and employs a Gaussian heatmap auxiliary branch to enhance model representation capability during training.To evaluate lane detection models,we rely on Fr´echet distance,parameterized F1-score,and additional synthetic metrics.Experimental results demonstrate that LDTR achieves state-of-the-art performance on well-known datasets.

关 键 词：TRANSFORMER lane detection anchor-chain 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程]