Papers › CANet: Curved Guide Line Network with Adaptive Decoder for Lane Detection

CANet: Curved Guide Line Network with Adaptive Decoder for Lane Detection

23 Apr 2023arXiv:2304.11546archive 2025-07-28

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

Lane detection is challenging due to the complicated on road scenarios and line deformation from different camera perspectives. Lots of solutions were proposed, but can not deal with corner lanes well. To address this problem, this paper proposes a new top-down deep learning lane detection approach, CANET. A lane instance is first responded by the heat-map on the U-shaped curved guide line at global semantic level, thus the corresponding features of each lane are aggregated at the response point. Then CANET obtains the heat-map response of the entire lane through conditional convolution, and finally decodes the point set to describe lanes via adaptive decoder. The experimental results show that CANET reaches SOTA in different metrics. Our code will be released soon.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DecoderLane Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Lane Detection CULane CANet-L(ResNet101) F1 score 79.86 #17 of 63 Archive leaderboard report
Lane Detection CULane CANet-M F1 score 79.16 #26 of 63 Archive leaderboard report
Lane Detection CULane CANet-S(ResNet18) F1 score 78.46 #31 of 63 Archive leaderboard report
Lane Detection CurveLanes CANet-L F1 score 87.87 #4 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-L Precision 91.69 #4 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-M F1 score 87.19 #6 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-M GFLOPs 22.6 #6 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-M Precision 91.53 #6 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-M Recall 83.25 #6 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-S F1 score 86.57 #7 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-S GFLOPs 13.1 #7 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-S Precision 91.37 #7 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-S Recall 82.25 #7 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-L(ResNet101) GFLOPs 45.7 #19 of 19 Archive leaderboard report
Lane Detection CurveLanes CANet-L(ResNet101) Recall 84.36 #19 of 19 Archive leaderboard report
Lane Detection TuSimple CANet-L(ResNet101) Accuracy 96.76% #10 of 43 Archive leaderboard report
Lane Detection TuSimple CANet-L(ResNet101) F1 score 97.77 #10 of 43 Archive leaderboard report
Lane Detection TuSimple CANet-M Accuracy 96.66% #11 of 43 Archive leaderboard report
Lane Detection TuSimple CANet-M F1 score 97.44 #11 of 43 Archive leaderboard report
Lane Detection TuSimple CANet-S Accuracy 96.56% #14 of 43 Archive leaderboard report
Lane Detection TuSimple CANet-S F1 score 97.51 #14 of 43 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections