{"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/how-robust-is-3d-human-pose-estimation-to","title":"How Robust is 3D Human Pose Estimation to Occlusion?","arxiv_id":"1808.09316","date":"2018-08-28","proceeding":null,"authors":["István Sárándi","Timm Linder","Kai O. Arras","Bastian Leibe"],"abstract":"Occlusion is commonplace in realistic human-robot shared environments, yet\nits effects are not considered in standard 3D human pose estimation benchmarks.\nThis leaves the question open: how robust are state-of-the-art 3D pose\nestimation methods against partial occlusions? We study several types of\nsynthetic occlusions over the Human3.6M dataset and find a method with\nstate-of-the-art benchmark performance to be sensitive even to low amounts of\nocclusion. Addressing this issue is key to progress in applications such as\ncollaborative and service robotics. We take a first step in this direction by\nimproving occlusion-robustness through training data augmentation with\nsynthetic occlusions. This also turns out to be an effective regularizer that\nis beneficial even for non-occluded test cases.","url_abs":"http://arxiv.org/abs/1808.09316v2","url_pdf":"http://arxiv.org/pdf/1808.09316v2.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":"how-robust-is-3d-human-pose-estimation-to","repo_url":"https://github.com/isarandi/synthetic-occlusion","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"how-robust-is-3d-human-pose-estimation-to","repo_url":"https://github.com/laurabragagnolo/human3.6m-occluded","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"3d-pose-estimation","task_name":"3D Pose Estimation"},{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.09316","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}