{"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/thin-slicing-network-a-deep-structured-model","title":"Thin-Slicing Network: A Deep Structured Model for Pose Estimation in Videos","arxiv_id":"1703.10898","date":"2017-03-31","proceeding":"CVPR 2017 7","authors":["Jie Song","Li-Min Wang","Luc van Gool","Otmar Hilliges"],"abstract":"Deep ConvNets have been shown to be effective for the task of human pose\nestimation from single images. However, several challenging issues arise in the\nvideo-based case such as self-occlusion, motion blur, and uncommon poses with\nfew or no examples in training data sets. Temporal information can provide\nadditional cues about the location of body joints and help to alleviate these\nissues. In this paper, we propose a deep structured model to estimate a\nsequence of human poses in unconstrained videos. This model can be efficiently\ntrained in an end-to-end manner and is capable of representing appearance of\nbody joints and their spatio-temporal relationships simultaneously. Domain\nknowledge about the human body is explicitly incorporated into the network\nproviding effective priors to regularize the skeletal structure and to enforce\ntemporal consistency. The proposed end-to-end architecture is evaluated on two\nwidely used benchmarks (Penn Action dataset and JHMDB dataset) for video-based\npose estimation. Our approach significantly outperforms the existing\nstate-of-the-art methods.","url_abs":"http://arxiv.org/abs/1703.10898v1","url_pdf":"http://arxiv.org/pdf/1703.10898v1.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":[],"tasks":[{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/pose-estimation-on-upenn-action","task":"Pose Estimation","dataset":"UPenn Action","model":"Thin-Slicing","rank_in_archive_order":4,"of":5,"metrics":{"Mean PCK@0.2":"96.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.10898","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}