{"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/crowdpose-efficient-crowded-scenes-pose","title":"CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmark","arxiv_id":"1812.00324","date":"2018-12-02","proceeding":"CVPR 2019 6","authors":["Jiefeng Li","Can Wang","Hao Zhu","Yihuan Mao","Hao-Shu Fang","Cewu Lu"],"abstract":"Multi-person pose estimation is fundamental to many computer vision tasks and\nhas made significant progress in recent years. However, few previous methods\nexplored the problem of pose estimation in crowded scenes while it remains\nchallenging and inevitable in many scenarios. Moreover, current benchmarks\ncannot provide an appropriate evaluation for such cases. In this paper, we\npropose a novel and efficient method to tackle the problem of pose estimation\nin the crowd and a new dataset to better evaluate algorithms. Our model\nconsists of two key components: joint-candidate single person pose estimation\n(SPPE) and global maximum joints association. With multi-peak prediction for\neach joint and global association using graph model, our method is robust to\ninevitable interference in crowded scenes and very efficient in inference. The\nproposed method surpasses the state-of-the-art methods on CrowdPose dataset by\n5.2 mAP and results on MSCOCO dataset demonstrate the generalization ability of\nour method. Source code and dataset will be made publicly available.","url_abs":"http://arxiv.org/abs/1812.00324v2","url_pdf":"http://arxiv.org/pdf/1812.00324v2.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":"crowdpose-efficient-crowded-scenes-pose","repo_url":"https://github.com/Jeff-sjtu/CrowdPose","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"crowdpose-efficient-crowded-scenes-pose","repo_url":"https://github.com/jeffffffli/CrowdPose","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"crowdpose-efficient-crowded-scenes-pose","repo_url":"https://github.com/laowang666888/ECSP1","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"crowdpose-efficient-crowded-scenes-pose","repo_url":"https://github.com/open-mmlab/mmpose","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"keypoint-detection","task_name":"Keypoint Detection"},{"task_slug":"multi-person-pose-estimation","task_name":"Multi-Person Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[{"slug":"crowdpose","name":"CrowdPose","full_name":"CrowdPose"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/multi-person-pose-estimation-on-crowdpose","task":"Multi-Person Pose Estimation","dataset":"CrowdPose","model":"Joint-candidate SPPE +","rank_in_archive_order":17,"of":28,"metrics":{"AP Easy":"75.5","AP Hard":"57.4","AP Medium":"66.3","FPS":"10.1","mAP @0.5:0.95":"66.0"},"uses_additional_data":false},{"leaderboard":"/sota/multi-person-pose-estimation-on-ochuman","task":"Multi-Person Pose Estimation","dataset":"OCHuman","model":"CrowdPose","rank_in_archive_order":6,"of":8,"metrics":{"AP50":"40.8","AP75":"29.9","Validation AP":"27.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.00324","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}