{"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/neural-kinematic-networks-for-unsupervised","title":"Neural Kinematic Networks for Unsupervised Motion Retargetting","arxiv_id":"1804.05653","date":"2018-04-16","proceeding":"CVPR 2018 6","authors":["Ruben Villegas","Jimei Yang","Duygu Ceylan","Honglak Lee"],"abstract":"We propose a recurrent neural network architecture with a Forward Kinematics\nlayer and cycle consistency based adversarial training objective for\nunsupervised motion retargetting. Our network captures the high-level\nproperties of an input motion by the forward kinematics layer, and adapts them\nto a target character with different skeleton bone lengths (e.g., shorter,\nlonger arms etc.). Collecting paired motion training sequences from different\ncharacters is expensive. Instead, our network utilizes cycle consistency to\nlearn to solve the Inverse Kinematics problem in an unsupervised manner. Our\nmethod works online, i.e., it adapts the motion sequence on-the-fly as new\nframes are received. In our experiments, we use the Mixamo animation data to\ntest our method for a variety of motions and characters and achieve\nstate-of-the-art results. We also demonstrate motion retargetting from\nmonocular human videos to 3D characters using an off-the-shelf 3D pose\nestimator.","url_abs":"http://arxiv.org/abs/1804.05653v1","url_pdf":"http://arxiv.org/pdf/1804.05653v1.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":"neural-kinematic-networks-for-unsupervised","repo_url":"https://github.com/rubenvillegas/cvpr2018nkn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1804.05653","atlas_url":"https://app.syntology.ai/?focus=1804.05653","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}