{"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/carlane-a-lane-detection-benchmark-for","title":"CARLANE: A Lane Detection Benchmark for Unsupervised Domain Adaptation from Simulation to multiple Real-World Domains","arxiv_id":"2206.08083","date":"2022-06-16","proceeding":null,"authors":["Julian Gebele","Bonifaz Stuhr","Johann Haselberger"],"abstract":"Unsupervised Domain Adaptation demonstrates great potential to mitigate domain shifts by transferring models from labeled source domains to unlabeled target domains. While Unsupervised Domain Adaptation has been applied to a wide variety of complex vision tasks, only few works focus on lane detection for autonomous driving. This can be attributed to the lack of publicly available datasets. To facilitate research in these directions, we propose CARLANE, a 3-way sim-to-real domain adaptation benchmark for 2D lane detection. CARLANE encompasses the single-target datasets MoLane and TuLane and the multi-target dataset MuLane. These datasets are built from three different domains, which cover diverse scenes and contain a total of 163K unique images, 118K of which are annotated. In addition we evaluate and report systematic baselines, including our own method, which builds upon Prototypical Cross-domain Self-supervised Learning. We find that false positive and false negative rates of the evaluated domain adaptation methods are high compared to those of fully supervised baselines. This affirms the need for benchmarks such as CARLANE to further strengthen research in Unsupervised Domain Adaptation for lane detection. CARLANE, all evaluated models and the corresponding implementations are publicly available at https://carlanebenchmark.github.io.","url_abs":"https://arxiv.org/abs/2206.08083v4","url_pdf":"https://arxiv.org/pdf/2206.08083v4.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":"carlane-a-lane-detection-benchmark-for","repo_url":"https://github.com/juliangebele/CARLANE","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"2d-semantic-segmentation","task_name":"2D Semantic Segmentation"},{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"lane-detection","task_name":"Lane Detection"},{"task_slug":"self-supervised-learning","task_name":"Self-Supervised Learning"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"unsupervised-domain-adaptation","task_name":"Unsupervised Domain Adaptation"},{"task_slug":"unsupervised-pre-training","task_name":"Unsupervised Pre-training"}],"methods":[{"method_slug":"sgpcs","method_name":"SGPCS"}],"datasets_introduced":[{"slug":"carlane-benchmark","name":"CARLANE Benchmark","full_name":""}],"methods_introduced":[{"slug":"sgpcs","name":"SGPCS","full_name":"Self-training Guided Prototypical Cross-domain Self-supervised learning"}],"results":[{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-SGPCS-ResNet18","rank_in_archive_order":1,"of":8,"metrics":{"Lane Accuracy (LA)":"93.94"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-SGADA-ResNet18","rank_in_archive_order":2,"of":8,"metrics":{"Lane Accuracy (LA)":"93.82"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-SGPCS-ResNet32","rank_in_archive_order":3,"of":8,"metrics":{"Lane Accuracy (LA)":"93.53"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-SGADA-ResNet32","rank_in_archive_order":4,"of":8,"metrics":{"Lane Accuracy (LA)":"93.31"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-ADDA-ResNet18","rank_in_archive_order":5,"of":8,"metrics":{"Lane Accuracy (LA)":"92.85"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-ADDA-ResNet32","rank_in_archive_order":6,"of":8,"metrics":{"Lane Accuracy (LA)":"92.39"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-DANN-ResNet32","rank_in_archive_order":7,"of":8,"metrics":{"Lane Accuracy (LA)":"90.91"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-molane","task":"Domain Adaptation","dataset":"MoLane","model":"UFLD-DANN-ResNet18","rank_in_archive_order":8,"of":8,"metrics":{"Lane Accuracy (LA)":"87.65"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-SGADA-ResNet32","rank_in_archive_order":1,"of":8,"metrics":{"Lane Accuracy (LA)":"91.63"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-SGPCS-ResNet18","rank_in_archive_order":2,"of":8,"metrics":{"Lane Accuracy (LA)":"91.57"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-SGPCS-ResNet32","rank_in_archive_order":3,"of":8,"metrics":{"Lane Accuracy (LA)":"91.55"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-SGADA-ResNet18","rank_in_archive_order":4,"of":8,"metrics":{"Lane Accuracy (LA)":"90.71"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-ADDA-ResNet32","rank_in_archive_order":5,"of":8,"metrics":{"Lane Accuracy (LA)":"90.22"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-ADDA-ResNet18","rank_in_archive_order":6,"of":8,"metrics":{"Lane Accuracy (LA)":"89.83"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-DANN-ResNet32","rank_in_archive_order":7,"of":8,"metrics":{"Lane Accuracy (LA)":"88.76"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-mulane","task":"Domain Adaptation","dataset":"MuLane","model":"UFLD-DANN-ResNet18","rank_in_archive_order":8,"of":8,"metrics":{"Lane Accuracy (LA)":"86.01"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-SGPCS-ResNet32","rank_in_archive_order":1,"of":8,"metrics":{"Lane Accuracy (LA)":"93.29"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-SGADA-ResNet32","rank_in_archive_order":2,"of":8,"metrics":{"Lane Accuracy (LA)":"92.04"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-SGADA-ResNet18","rank_in_archive_order":3,"of":8,"metrics":{"Lane Accuracy (LA)":"91.70"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-SGPCS-ResNet18","rank_in_archive_order":4,"of":8,"metrics":{"Lane Accuracy (LA)":"91.55"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-ADDA-ResNet32","rank_in_archive_order":5,"of":8,"metrics":{"Lane Accuracy (LA)":"91.39"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-DANN-ResNet32","rank_in_archive_order":6,"of":8,"metrics":{"Lane Accuracy (LA)":"91.06"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-ADDA-ResNet18","rank_in_archive_order":7,"of":8,"metrics":{"Lane Accuracy (LA)":"90.72"},"uses_additional_data":false},{"leaderboard":"/sota/domain-adaptation-on-tulane","task":"Domain Adaptation","dataset":"TuLane","model":"UFLD-DANN-ResNet18","rank_in_archive_order":8,"of":8,"metrics":{"Lane Accuracy (LA)":"88.74"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2206.08083","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}