{"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/weakly-and-semi-supervised-human-body-part","title":"Weakly and Semi Supervised Human Body Part Parsing via Pose-Guided Knowledge Transfer","arxiv_id":"1805.04310","date":"2018-05-11","proceeding":"CVPR 2018 6","authors":["Hao-Shu Fang","Guansong Lu","Xiaolin Fang","Jianwen Xie","Yu-Wing Tai","Cewu Lu"],"abstract":"Human body part parsing, or human semantic part segmentation, is fundamental\nto many computer vision tasks. In conventional semantic segmentation methods,\nthe ground truth segmentations are provided, and fully convolutional networks\n(FCN) are trained in an end-to-end scheme. Although these methods have\ndemonstrated impressive results, their performance highly depends on the\nquantity and quality of training data. In this paper, we present a novel method\nto generate synthetic human part segmentation data using easily-obtained human\nkeypoint annotations. Our key idea is to exploit the anatomical similarity\namong human to transfer the parsing results of a person to another person with\nsimilar pose. Using these estimated results as additional training data, our\nsemi-supervised model outperforms its strong-supervised counterpart by 6 mIOU\non the PASCAL-Person-Part dataset, and we achieve state-of-the-art human\nparsing results. Our approach is general and can be readily extended to other\nobject/animal parsing task assuming that their anatomical similarity can be\nannotated by keypoints. The proposed model and accompanying source code are\navailable at https://github.com/MVIG-SJTU/WSHP","url_abs":"http://arxiv.org/abs/1805.04310v1","url_pdf":"http://arxiv.org/pdf/1805.04310v1.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":"weakly-and-semi-supervised-human-body-part","repo_url":"https://github.com/MVIG-SJTU/WSHP","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"human-parsing","task_name":"Human Parsing"},{"task_slug":"human-part-segmentation","task_name":"Human Part Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/human-part-segmentation-on-pascal-person-part","task":"Human Part Segmentation","dataset":"PASCAL-Part","model":"WSHP","rank_in_archive_order":3,"of":7,"metrics":{"mIoU":"67.60"},"uses_additional_data":true}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.04310","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}