{"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/mt-vae-learning-motion-transformations-to","title":"MT-VAE: Learning Motion Transformations to Generate Multimodal Human Dynamics","arxiv_id":"1808.04545","date":"2018-08-14","proceeding":"ECCV 2018 9","authors":["Xinchen Yan","Akash Rastogi","Ruben Villegas","Kalyan Sunkavalli","Eli Shechtman","Sunil Hadap","Ersin Yumer","Honglak Lee"],"abstract":"Long-term human motion can be represented as a series of motion\nmodes---motion sequences that capture short-term temporal dynamics---with\ntransitions between them. We leverage this structure and present a novel Motion\nTransformation Variational Auto-Encoders (MT-VAE) for learning motion sequence\ngeneration. Our model jointly learns a feature embedding for motion modes (that\nthe motion sequence can be reconstructed from) and a feature transformation\nthat represents the transition of one motion mode to the next motion mode. Our\nmodel is able to generate multiple diverse and plausible motion sequences in\nthe future from the same input. We apply our approach to both facial and full\nbody motion, and demonstrate applications like analogy-based motion transfer\nand video synthesis.","url_abs":"http://arxiv.org/abs/1808.04545v1","url_pdf":"http://arxiv.org/pdf/1808.04545v1.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":"mt-vae-learning-motion-transformations-to","repo_url":"https://github.com/xcyan/eccv18_mtvae","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"human-dynamics","task_name":"Human Dynamics"},{"task_slug":"human-pose-forecasting","task_name":"Human Pose Forecasting"},{"task_slug":"motion-prediction","task_name":"motion prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/human-pose-forecasting-on-human36m","task":"Human Pose Forecasting","dataset":"Human3.6M","model":"MT-VAE","rank_in_archive_order":33,"of":33,"metrics":{"ADE":"457","APD":"403","FDE":"595","MMADE":"716","MMFDE":"883"},"uses_additional_data":false},{"leaderboard":"/sota/human-pose-forecasting-on-humaneva-i","task":"Human Pose Forecasting","dataset":"HumanEva-I","model":"MT-VAE","rank_in_archive_order":11,"of":11,"metrics":{"ADE@2000ms":"345","APD@2000ms":"21","FDE@2000ms":"403"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.04545","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}