CLRerNet: Improving Confidence of Lane Detection with LaneIoU

15 May 2023  ยท  Hiroto Honda, Yusuke Uchida ยท

Lane marker detection is a crucial component of the autonomous driving and driver assistance systems. Modern deep lane detection methods with row-based lane representation exhibit excellent performance on lane detection benchmarks. Through preliminary oracle experiments, we firstly disentangle the lane representation components to determine the direction of our approach. We show that correct lane positions are already among the predictions of an existing row-based detector, and the confidence scores that accurately represent intersection-over-union (IoU) with ground truths are the most beneficial. Based on the finding, we propose LaneIoU that better correlates with the metric, by taking the local lane angles into consideration. We develop a novel detector coined CLRerNet featuring LaneIoU for the target assignment cost and loss functions aiming at the improved quality of confidence scores. Through careful and fair benchmark including cross validation, we demonstrate that CLRerNet outperforms the state-of-the-art by a large margin - enjoying F1 score of 81.43% compared with 80.47% of the existing method on CULane, and 86.47% compared with 86.10% on CurveLanes.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Lane Detection CULane CLRerNet-DLA34 F1 score 81.12 # 1
Lane Detection CULane CLRerNet-Res101 F1 score 80.91 # 2
Lane Detection CULane CLRerNet-Res34 F1 score 80.76 # 4
Lane Detection CurveLanes CLRerNet-DLA34 F1 score 86.47 # 7
Precision 91.66 # 4
Recall 81.83 # 7
GFLOPs 18.4 # 7
Lane Detection CurveLanes CLRNet-DLA34 F1 score 86.1 # 8
Precision 91.4 # 6
Recall 81.39 # 8
GFLOPs 18.4 # 7

Methods


No methods listed for this paper. Add relevant methods here