{"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/a-richly-annotated-dataset-for-pedestrian","title":"A Richly Annotated Dataset for Pedestrian Attribute Recognition","arxiv_id":"1603.07054","date":"2016-03-23","proceeding":null,"authors":["Dangwei Li","Zhang Zhang","Xiaotang Chen","Haibin Ling","Kaiqi Huang"],"abstract":"In this paper, we aim to improve the dataset foundation for pedestrian\nattribute recognition in real surveillance scenarios. Recognition of human\nattributes, such as gender, and clothes types, has great prospects in real\napplications. However, the development of suitable benchmark datasets for\nattribute recognition remains lagged behind. Existing human attribute datasets\nare collected from various sources or an integration of pedestrian\nre-identification datasets. Such heterogeneous collection poses a big challenge\non developing high quality fine-grained attribute recognition algorithms.\nFurthermore, human attribute recognition are generally severely affected by\nenvironmental or contextual factors, such as viewpoints, occlusions and body\nparts, while existing attribute datasets barely care about them. To tackle\nthese problems, we build a Richly Annotated Pedestrian (RAP) dataset from real\nmulti-camera surveillance scenarios with long term collection, where data\nsamples are annotated with not only fine-grained human attributes but also\nenvironmental and contextual factors. RAP has in total 41,585 pedestrian\nsamples, each of which is annotated with 72 attributes as well as viewpoints,\nocclusions, body parts information. To our knowledge, the RAP dataset is the\nlargest pedestrian attribute dataset, which is expected to greatly promote the\nstudy of large-scale attribute recognition systems. Furthermore, we empirically\nanalyze the effects of different environmental and contextual factors on\npedestrian attribute recognition. Experimental results demonstrate that\nviewpoints, occlusions and body parts information could assist attribute\nrecognition a lot in real applications.","url_abs":"http://arxiv.org/abs/1603.07054v3","url_pdf":"http://arxiv.org/pdf/1603.07054v3.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":"a-richly-annotated-dataset-for-pedestrian","repo_url":"https://github.com/ajithvallabai/Pedestrian_Attribute_Recognition","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"a-richly-annotated-dataset-for-pedestrian","repo_url":"https://github.com/guotengg/sspnet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"a-richly-annotated-dataset-for-pedestrian","repo_url":"https://github.com/valencebond/Rethinking_of_PAR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"pedestrian-attribute-recognition","task_name":"Pedestrian Attribute Recognition"}],"methods":[],"datasets_introduced":[{"slug":"rap","name":"RAP","full_name":"Richly Annotated Pedestrian"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1603.07054","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}