{"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/the-cityscapes-dataset-for-semantic-urban","title":"The Cityscapes Dataset for Semantic Urban Scene Understanding","arxiv_id":"1604.01685","date":"2016-04-06","proceeding":"CVPR 2016 6","authors":["Marius Cordts","Mohamed Omran","Sebastian Ramos","Timo Rehfeld","Markus Enzweiler","Rodrigo Benenson","Uwe Franke","Stefan Roth","Bernt Schiele"],"abstract":"Visual understanding of complex urban street scenes is an enabling factor for\na wide range of applications. Object detection has benefited enormously from\nlarge-scale datasets, especially in the context of deep learning. For semantic\nurban scene understanding, however, no current dataset adequately captures the\ncomplexity of real-world urban scenes.\n  To address this, we introduce Cityscapes, a benchmark suite and large-scale\ndataset to train and test approaches for pixel-level and instance-level\nsemantic labeling. Cityscapes is comprised of a large, diverse set of stereo\nvideo sequences recorded in streets from 50 different cities. 5000 of these\nimages have high quality pixel-level annotations; 20000 additional images have\ncoarse annotations to enable methods that leverage large volumes of\nweakly-labeled data. Crucially, our effort exceeds previous attempts in terms\nof dataset size, annotation richness, scene variability, and complexity. Our\naccompanying empirical study provides an in-depth analysis of the dataset\ncharacteristics, as well as a performance evaluation of several\nstate-of-the-art approaches based on our benchmark.","url_abs":"http://arxiv.org/abs/1604.01685v2","url_pdf":"http://arxiv.org/pdf/1604.01685v2.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":"the-cityscapes-dataset-for-semantic-urban","repo_url":"https://github.com/Ivan-LZY/SG-Cyclingscapes","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"the-cityscapes-dataset-for-semantic-urban","repo_url":"https://github.com/gjp1203/LIV360SV","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"the-cityscapes-dataset-for-semantic-urban","repo_url":"https://github.com/valeoai/VideoActionModel","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"scene-understanding","task_name":"Scene Understanding"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[{"slug":"cityscapes","name":"Cityscapes","full_name":""},{"slug":"cityscapes-seq","name":"Cityscapes-Seq","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1604.01685","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}