{"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/unifying-part-detection-and-association-for","title":"Unifying Part Detection and Association for Recurrent Multi-Person Pose Estimation","arxiv_id":"1904.11864","date":"2019-04-26","proceeding":null,"authors":["Rania Briq","Andreas Doering","Juergen Gall"],"abstract":"We propose a joint model of human joint detection and association for 2D\nmulti-person pose estimation (MPPE). The approach unifies training of joint\ndetection and association without a need for further processing or\nsophisticated heuristics in order to associate the joints with people\nindividually. The approach consists of two stages, where in the first stage\njoint detection heatmaps and association features are extracted, and in the\nsecond stage, whose input are the extracted features of the first stage, we\nintroduce a recurrent neural network (RNN) which predicts the heatmaps of a\nsingle person's joints in each iteration. In addition, the network learns a\nstopping criterion in order to halt once it has identified all individuals in\nthe image. This approach allowed us to eliminate several heuristic assumptions\nand parameters needed for association which do not necessarily hold true.\nAdditionally, such an end-to-end approach allows the final objective to be\nknown and directly optimized over during training. We evaluated our model on\nthe challenging MSCOCO dataset and obtained an improvement over the baseline,\nparticularly in challenging scenes with occlusions.","url_abs":"http://arxiv.org/abs/1904.11864v1","url_pdf":"http://arxiv.org/pdf/1904.11864v1.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":"unifying-part-detection-and-association-for","repo_url":"https://github.com/briqr/end2end_humanpose_estimation","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"multi-person-pose-estimation","task_name":"Multi-Person Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}