{"url":"/sota/3d-lane-detection-on-openlane","task":{"name":"3D Lane Detection","url":"/task/3d-lane-detection","note":null},"dataset":{"name":"OpenLane","url":"/dataset/openlane"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"The goal of **3D Lane Detection** is to perceive lanes that provide guidance for autonomous vehicles. A lane can be represented as a visible laneline or a conceptual centerline. Furthermore, a lane obtains extra attributes from the understanding of the surrounding environment.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [OpenLane-V2](https://github.com/OpenDriveLab/OpenLane-V2 ) )</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["F1 (all)","Up & Down","Curve","Extreme Weather","Night","Intersection","Merge & Split","FPS (pytorch)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"F1 (all)":"higher","Up & Down":null,"Curve":null,"Extreme Weather":null,"Night":null,"Intersection":null,"Merge & Split":null,"FPS (pytorch)":null}},"counts":{"rows":21,"rows_with_code":12,"rows_with_paper_page":21,"rows_dated":21,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"GLane3D(Swin-B)","metrics":{"Curve":"72.7","Extreme Weather":"63.8","F1 (all)":"66.0","Intersection":"57.9","Merge & Split":"67.7","Night":"62.0","Up & Down":"61.7"},"uses_additional_data":false,"paper_date":"2025-03-31","paper":"/paper/glane3d-detecting-lanes-with-graph-of-3d","paper_url":"https://arxiv.org/abs/2503.23882v1","paper_title":"GLane3D : Detecting Lanes with Graph of 3D Keypoints","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"GLane3D(ResNet-50)","metrics":{"Curve":"71.1","Extreme Weather":"60.1","F1 (all)":"63.9","Intersection":"55.0","Merge & Split":"64.8","Night":"60.2","Up & Down":"58.2"},"uses_additional_data":false,"paper_date":"2025-03-31","paper":"/paper/glane3d-detecting-lanes-with-graph-of-3d","paper_url":"https://arxiv.org/abs/2503.23882v1","paper_title":"GLane3D : Detecting Lanes with Graph of 3D Keypoints","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"PVALane (Swin-B)","metrics":{"Curve":"67.7","Extreme Weather":"64.0","F1 (all)":"63.4","Intersection":"53.6","Merge & Split":"60.8","Night":"58.6","Up & Down":"56.1"},"uses_additional_data":false,"paper_date":"2024-03-24","paper":"/paper/pvalane-prior-guided-3d-lane-detection-with","paper_url":"https://ojs.aaai.org/index.php/AAAI/article/view/28592","paper_title":"PVALane: Prior-Guided 3D Lane Detection with View-Agnostic Feature Alignment","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"PVALane (ResNet-50)","metrics":{"Curve":"67.3","Extreme Weather":"62.0","F1 (all)":"62.7","Intersection":"53.4","Merge & Split":"60.0","Night":"57.2","Up & Down":"54.1"},"uses_additional_data":false,"paper_date":"2024-03-24","paper":"/paper/pvalane-prior-guided-3d-lane-detection-with","paper_url":"https://ojs.aaai.org/index.php/AAAI/article/view/28592","paper_title":"PVALane: Prior-Guided 3D Lane Detection with View-Agnostic Feature Alignment","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"LATR","metrics":{"Curve":"68.2","Extreme Weather":"57.1","F1 (all)":"61.9","Intersection":"52.3","Merge & Split":"61.5","Night":"55.4","Up & Down":"55.2"},"uses_additional_data":false,"paper_date":"2023-08-08","paper":"/paper/latr-3d-lane-detection-from-monocular-images","paper_url":"https://arxiv.org/abs/2308.04583v2","paper_title":"LATR: 3D Lane Detection from Monocular Images with Transformer","code":"https://github.com/jmoonr/latr","n_code_links":1,"syntology":{"n_ran":11,"n_unverified":3,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"RFTR (ResNet-50)","metrics":{"Curve":"-","Extreme Weather":"-","F1 (all)":"61.8","Intersection":"-","Merge & Split":"-","Night":"-","Up & Down":"-"},"uses_additional_data":false,"paper_date":"2024-07-15","paper":"/paper/repvf-a-unified-vector-fields-representation","paper_url":"https://arxiv.org/abs/2407.10876v2","paper_title":"RepVF: A Unified Vector Fields Representation for Multi-task 3D Perception","code":"https://github.com/jbji/repvf","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":2,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"PVALane (ResNet-18)","metrics":{"Curve":"65.7","Extreme Weather":"59.5","F1 (all)":"61.2","Intersection":"52.2","Merge & Split":"58.7","Night":"56.5","Up & Down":"52.6"},"uses_additional_data":false,"paper_date":"2024-03-24","paper":"/paper/pvalane-prior-guided-3d-lane-detection-with","paper_url":"https://ojs.aaai.org/index.php/AAAI/article/view/28592","paper_title":"PVALane: Prior-Guided 3D Lane Detection with View-Agnostic Feature Alignment","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"PETRv2-V∗ (VoVNetV2 with 400 anchor points)","metrics":{"Curve":"-","Extreme Weather":"-","F1 (all)":"61.2","FPS (pytorch)":"-","Intersection":"-","Merge & Split":"-","Night":"-","Up & Down":"-"},"uses_additional_data":false,"paper_date":"2022-06-02","paper":"/paper/petrv2-a-unified-framework-for-3d-perception","paper_url":"https://arxiv.org/abs/2206.01256v3","paper_title":"PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images","code":"https://github.com/megvii-research/petr","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":9,"model":"LaneCPP","metrics":{"Curve":"64.4","Extreme Weather":"56.7","F1 (all)":"60.3","Intersection":"52.0","Merge & Split":"58.7","Night":"54.9","Up & Down":"53.6"},"uses_additional_data":false,"paper_date":"2024-06-12","paper":"/paper/lanecpp-continuous-3d-lane-detection-using-1","paper_url":"https://arxiv.org/abs/2406.08381v1","paper_title":"LaneCPP: Continuous 3D Lane Detection using Physical Priors","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":10,"model":"BEV-LaneDet","metrics":{"Curve":"63.1","Extreme Weather":"53.4","F1 (all)":"58.4","FPS (pytorch)":"102","Intersection":"50.3","Merge & Split":"53.7","Night":"53.4","Up & Down":"48.7"},"uses_additional_data":false,"paper_date":"2022-10-12","paper":"/paper/bev-lane-det-fast-lane-detection-on-bev","paper_url":"https://arxiv.org/abs/2210.06006v3","paper_title":"BEV-LaneDet: a Simple and Effective 3D Lane Detection Baseline","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":11,"model":"PETRv2-V (VoVNetV2)","metrics":{"Curve":"-","Extreme Weather":"-","F1 (all)":"57.8","FPS (pytorch)":"-","Intersection":"-","Merge & Split":"-","Night":"-","Up & Down":"-"},"uses_additional_data":false,"paper_date":"2022-06-02","paper":"/paper/petrv2-a-unified-framework-for-3d-perception","paper_url":"https://arxiv.org/abs/2206.01256v3","paper_title":"PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images","code":"https://github.com/megvii-research/petr","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":12,"model":"M^2-3DLaneNet (Camera + Lidar)","metrics":{"Curve":"60.7","Extreme Weather":"56.2","F1 (all)":"55.5","FPS (pytorch)":"-","Intersection":"43.8","Merge & Split":"51.4","Night":"51.6","Up & Down":"53.4"},"uses_additional_data":false,"paper_date":"2022-09-13","paper":"/paper/m-2-3dlanenet-multi-modal-3d-lane-detection","paper_url":"https://arxiv.org/abs/2209.05996v3","paper_title":"M$^2$-3DLaneNet: Exploring Multi-Modal 3D Lane Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"Anchor3DLane-T† (multi-frame + iterative regression)","metrics":{"Curve":"58.0","Extreme Weather":"52.7","F1 (all)":"54.3","FPS (pytorch)":"-","Intersection":"45.8","Merge & Split":"51.7","Night":"48.7","Up & Down":"47.2"},"uses_additional_data":false,"paper_date":"2023-01-06","paper":"/paper/anchor3dlane-learning-to-regress-3d-anchors","paper_url":"https://arxiv.org/abs/2301.02371v2","paper_title":"Anchor3DLane: Learning to Regress 3D Anchors for Monocular 3D Lane Detection","code":"https://github.com/tusen-ai/anchor3dlane","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"Anchor3DLane† (iterative regression)","metrics":{"Curve":"57.2","Extreme Weather":"52.5","F1 (all)":"53.7","Intersection":"45.4","Merge & Split":"51.2","Night":"47.8","Up & Down":"46.7"},"uses_additional_data":false,"paper_date":"2023-01-06","paper":"/paper/anchor3dlane-learning-to-regress-3d-anchors","paper_url":"https://arxiv.org/abs/2301.02371v2","paper_title":"Anchor3DLane: Learning to Regress 3D Anchors for Monocular 3D Lane Detection","code":"https://github.com/tusen-ai/anchor3dlane","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"Anchor3DLane (ResNet-18)","metrics":{"Curve":"56.2","Extreme Weather":"51.9 ","F1 (all)":"53.1","Intersection":"44.2","Merge & Split":"50.5","Night":"47.2","Up & Down":"45.5"},"uses_additional_data":false,"paper_date":"2023-01-06","paper":"/paper/anchor3dlane-learning-to-regress-3d-anchors","paper_url":"https://arxiv.org/abs/2301.02371v2","paper_title":"Anchor3DLane: Learning to Regress 3D Anchors for Monocular 3D Lane Detection","code":"https://github.com/tusen-ai/anchor3dlane","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"PersFormer (version 1.2)","metrics":{"Curve":"58.4","Extreme Weather":"51.8","F1 (all)":"52.9","FPS (pytorch)":"-","Intersection":"42.1","Merge & Split":"50.9","Night":"47.4","Up & Down":"47.5"},"uses_additional_data":false,"paper_date":"2022-03-21","paper":"/paper/persformer-3d-lane-detection-via-perspective-1","paper_url":"https://arxiv.org/abs/2203.11089v3","paper_title":"PersFormer: 3D Lane Detection via Perspective Transformer and the OpenLane Benchmark","code":"https://github.com/OpenDriveLab/OpenLane","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":7,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"PETRv2-E (EfficientNet)","metrics":{"Curve":"-","Extreme Weather":"-","F1 (all)":"51.9","FPS (pytorch)":"-","Intersection":"-","Merge & Split":"-","Night":"-","Up & Down":"-"},"uses_additional_data":false,"paper_date":"2022-06-02","paper":"/paper/petrv2-a-unified-framework-for-3d-perception","paper_url":"https://arxiv.org/abs/2206.01256v3","paper_title":"PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images","code":"https://github.com/megvii-research/petr","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":18,"model":"PersFormer (version 1.1)","metrics":{"Curve":"58.7","Extreme Weather":"54.0","F1 (all)":"50.5","FPS (pytorch)":"-","Intersection":"41.6","Merge & Split":"53.1","Night":"50.0","Up & Down":"45.6"},"uses_additional_data":false,"paper_date":"2022-03-21","paper":"/paper/persformer-3d-lane-detection-via-perspective-1","paper_url":"https://arxiv.org/abs/2203.11089v3","paper_title":"PersFormer: 3D Lane Detection via Perspective Transformer and the OpenLane Benchmark","code":"https://github.com/OpenDriveLab/OpenLane","n_code_links":2,"syntology":{"n_ran":8,"n_unverified":7,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"CurveFormer","metrics":{"Curve":"56.6","Extreme Weather":"49.7","F1 (all)":"50.5","FPS (pytorch)":"-","Intersection":"42.9","Merge & Split":"45.4","Night":"49.1","Up & Down":"45.2"},"uses_additional_data":false,"paper_date":"2022-09-16","paper":"/paper/curveformer-3d-lane-detection-by-curve","paper_url":"https://arxiv.org/abs/2209.07989v2","paper_title":"CurveFormer: 3D Lane Detection by Curve Propagation with Curve Queries and Attention","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"3D-LaneNet","metrics":{"Curve":"46.5","Extreme Weather":"47.5","F1 (all)":"44.1","FPS (pytorch)":"-","Intersection":"32.1","Merge & Split":"41.7","Night":"41.5","Up & Down":"40.8"},"uses_additional_data":false,"paper_date":"2018-11-26","paper":"/paper/3d-lanenet-end-to-end-3d-multiple-lane","paper_url":"https://arxiv.org/abs/1811.10203v3","paper_title":"3D-LaneNet: End-to-End 3D Multiple Lane Detection","code":"https://github.com/yuliangguo/Pytorch_Generalized_3D_Lane_Detection","n_code_links":1,"syntology":null},{"rank_in_archive_order":21,"model":"Gen-LaneNet","metrics":{"Curve":"33.5","Extreme Weather":"28.1","F1 (all)":"32.3","FPS (pytorch)":"-","Intersection":"21.4","Merge & Split":"31.0","Night":"18.7","Up & Down":"25.4"},"uses_additional_data":false,"paper_date":"2020-03-24","paper":"/paper/gen-lanenet-a-generalized-and-scalable","paper_url":"https://arxiv.org/abs/2003.10656v1","paper_title":"Gen-LaneNet: A Generalized and Scalable Approach for 3D Lane Detection","code":"https://github.com/yuliangguo/Pytorch_Generalized_3D_Lane_Detection","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":7,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":24,"n_unverified":12,"n_samples":36,"n_pointer_only_licence":1,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":34,"n_unverified":19,"n_samples":53,"n_pointer_only_licence":3,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}