{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/lanenet-real-time-lane-detection-networks-for","title":"LaneNet: Real-Time Lane Detection Networks for Autonomous Driving","arxiv_id":"1807.01726","date":"2018-07-04","proceeding":null,"authors":["Ze Wang","Weiqiang Ren","Qiang Qiu"],"abstract":"Lane detection is to detect lanes on the road and provide the accurate\nlocation and shape of each lane. It severs as one of the key techniques to\nenable modern assisted and autonomous driving systems. However, several unique\nproperties of lanes challenge the detection methods. The lack of distinctive\nfeatures makes lane detection algorithms tend to be confused by other objects\nwith similar local appearance. Moreover, the inconsistent number of lanes on a\nroad as well as diverse lane line patterns, e.g. solid, broken, single, double,\nmerging, and splitting lines further hamper the performance. In this paper, we\npropose a deep neural network based method, named LaneNet, to break down the\nlane detection into two stages: lane edge proposal and lane line localization.\nStage one uses a lane edge proposal network for pixel-wise lane edge\nclassification, and the lane line localization network in stage two then\ndetects lane lines based on lane edge proposals. Please note that the goal of\nour LaneNet is built to detect lane line only, which introduces more\ndifficulties on suppressing the false detections on the similar lane marks on\nthe road like arrows and characters. Despite all the difficulties, our lane\ndetection is shown to be robust to both highway and urban road scenarios method\nwithout relying on any assumptions on the lane number or the lane line\npatterns. The high running speed and low computational cost endow our LaneNet\nthe capability of being deployed on vehicle-based systems. Experiments validate\nthat our LaneNet consistently delivers outstanding performances on real world\ntraffic scenarios.","url_abs":"http://arxiv.org/abs/1807.01726v1","url_pdf":"http://arxiv.org/pdf/1807.01726v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"lanenet-real-time-lane-detection-networks-for","repo_url":"https://github.com/klintan/pytorch-lanenet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"lanenet-real-time-lane-detection-networks-for","repo_url":"https://github.com/IvanVassi/LaneNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"edge-classification","task_name":"Edge Classification"},{"task_slug":"lane-detection","task_name":"Lane Detection"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.01726","atlas_url":"https://app.syntology.ai/?focus=1807.01726","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}