{"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/quaternet-a-quaternion-based-recurrent-model","title":"QuaterNet: A Quaternion-based Recurrent Model for Human Motion","arxiv_id":"1805.06485","date":"2018-05-16","proceeding":null,"authors":["Dario Pavllo","David Grangier","Michael Auli"],"abstract":"Deep learning for predicting or generating 3D human pose sequences is an\nactive research area. Previous work regresses either joint rotations or joint\npositions. The former strategy is prone to error accumulation along the\nkinematic chain, as well as discontinuities when using Euler angle or\nexponential map parameterizations. The latter requires re-projection onto\nskeleton constraints to avoid bone stretching and invalid configurations. This\nwork addresses both limitations. Our recurrent network, QuaterNet, represents\nrotations with quaternions and our loss function performs forward kinematics on\na skeleton to penalize absolute position errors instead of angle errors. On\nshort-term predictions, QuaterNet improves the state-of-the-art quantitatively.\nFor long-term generation, our approach is qualitatively judged as realistic as\nrecent neural strategies from the graphics literature.","url_abs":"http://arxiv.org/abs/1805.06485v2","url_pdf":"http://arxiv.org/pdf/1805.06485v2.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":"quaternet-a-quaternion-based-recurrent-model","repo_url":"https://github.com/facebookresearch/QuaterNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"motion-estimation","task_name":"Motion Estimation"},{"task_slug":null,"task_name":"Position"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1805.06485","atlas_url":"https://app.syntology.ai/?focus=1805.06485","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}