{"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/citypersons-a-diverse-dataset-for-pedestrian","title":"CityPersons: A Diverse Dataset for Pedestrian Detection","arxiv_id":"1702.05693","date":"2017-02-19","proceeding":"CVPR 2017 7","authors":["Shanshan Zhang","Rodrigo Benenson","Bernt Schiele"],"abstract":"Convnets have enabled significant progress in pedestrian detection recently,\nbut there are still open questions regarding suitable architectures and\ntraining data. We revisit CNN design and point out key adaptations, enabling\nplain FasterRCNN to obtain state-of-the-art results on the Caltech dataset.\n  To achieve further improvement from more and better data, we introduce\nCityPersons, a new set of person annotations on top of the Cityscapes dataset.\nThe diversity of CityPersons allows us for the first time to train one single\nCNN model that generalizes well over multiple benchmarks. Moreover, with\nadditional training with CityPersons, we obtain top results using FasterRCNN on\nCaltech, improving especially for more difficult cases (heavy occlusion and\nsmall scale) and providing higher localization quality.","url_abs":"http://arxiv.org/abs/1702.05693v1","url_pdf":"http://arxiv.org/pdf/1702.05693v1.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":"citypersons-a-diverse-dataset-for-pedestrian","repo_url":"https://github.com/aibeedetect/bfjdet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"citypersons-a-diverse-dataset-for-pedestrian","repo_url":"https://github.com/hnuzhy/bpjdet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"pedestrian-detection","task_name":"Pedestrian Detection"}],"methods":[],"datasets_introduced":[{"slug":"citypersons","name":"CityPersons","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/pedestrian-detection-on-caltech","task":"Pedestrian Detection","dataset":"Caltech","model":"Zhang et al. *","rank_in_archive_order":14,"of":33,"metrics":{"Reasonable Miss Rate":"5.1"},"uses_additional_data":false},{"leaderboard":"/sota/pedestrian-detection-on-caltech","task":"Pedestrian Detection","dataset":"Caltech","model":"Zhang et al.","rank_in_archive_order":16,"of":33,"metrics":{"Reasonable Miss Rate":"5.8"},"uses_additional_data":false},{"leaderboard":"/sota/pedestrian-detection-on-citypersons","task":"Pedestrian Detection","dataset":"CityPersons","model":"FRCNN+Seg","rank_in_archive_order":19,"of":22,"metrics":{"Large MR^-2":"8.0","Medium MR^-2":"6.7","Reasonable MR^-2":"14.8","Small MR^-2":"22.6"},"uses_additional_data":false},{"leaderboard":"/sota/pedestrian-detection-on-citypersons","task":"Pedestrian Detection","dataset":"CityPersons","model":"FRCNN","rank_in_archive_order":20,"of":22,"metrics":{"Large MR^-2":"7.9","Medium MR^-2":"7.2","Reasonable MR^-2":"15.4","Small MR^-2":"25.6"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1702.05693","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}