{"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/human-motion-diffusion-model","title":"Human Motion Diffusion Model","arxiv_id":"2209.14916","date":"2022-09-29","proceeding":null,"authors":["Guy Tevet","Sigal Raab","Brian Gordon","Yonatan Shafir","Daniel Cohen-Or","Amit H. Bermano"],"abstract":"Natural and expressive human motion generation is the holy grail of computer animation. It is a challenging task, due to the diversity of possible motion, human perceptual sensitivity to it, and the difficulty of accurately describing it. Therefore, current generative solutions are either low-quality or limited in expressiveness. Diffusion models, which have already shown remarkable generative capabilities in other domains, are promising candidates for human motion due to their many-to-many nature, but they tend to be resource hungry and hard to control. In this paper, we introduce Motion Diffusion Model (MDM), a carefully adapted classifier-free diffusion-based generative model for the human motion domain. MDM is transformer-based, combining insights from motion generation literature. A notable design-choice is the prediction of the sample, rather than the noise, in each diffusion step. This facilitates the use of established geometric losses on the locations and velocities of the motion, such as the foot contact loss. As we demonstrate, MDM is a generic approach, enabling different modes of conditioning, and different generation tasks. We show that our model is trained with lightweight resources and yet achieves state-of-the-art results on leading benchmarks for text-to-motion and action-to-motion. https://guytevet.github.io/mdm-page/ .","url_abs":"https://arxiv.org/abs/2209.14916v2","url_pdf":"https://arxiv.org/pdf/2209.14916v2.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":"human-motion-diffusion-model","repo_url":"https://github.com/guytevet/motion-diffusion-model","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"3d-generation","task_name":"3D Generation"},{"task_slug":"motion-generation","task_name":"Motion Generation"},{"task_slug":"motion-synthesis","task_name":"Motion Synthesis"},{"task_slug":"model","task_name":"model"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-generation-on-e-t-the-exceptional","task":"3D Generation","dataset":"E.T. the Exceptional Trajectories","model":"MDM","rank_in_archive_order":4,"of":4,"metrics":{"ClaTr-Score":"18.32 ","Classifier-F1":"0.34","FD_ClaTr":"6.79 "},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-humanact12","task":"Motion Synthesis","dataset":"HumanAct12","model":"MDM","rank_in_archive_order":1,"of":2,"metrics":{"Accuracy":"0.99","FID":"0.08","Multimodality":"2.58"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-humanml3d","task":"Motion Synthesis","dataset":"HumanML3D","model":"MDM","rank_in_archive_order":29,"of":37,"metrics":{"Diversity":"9.559","FID":"0.544","Multimodality":"2.799","R Precision Top3":"0.611"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-inter-x","task":"Motion Synthesis","dataset":"Inter-X","model":"MDM","rank_in_archive_order":4,"of":6,"metrics":{"FID":"23.701","MMDist":"9.548","MModality":"3.490","R-Precision Top3":"0.426"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-interhuman","task":"Motion Synthesis","dataset":"InterHuman","model":"MDM","rank_in_archive_order":8,"of":10,"metrics":{"FID":"9.167","MMDist":"7.125","MModality":"2.35","R-Precision Top3":"0.339"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-kit-motion-language","task":"Motion Synthesis","dataset":"KIT Motion-Language","model":"MDM","rank_in_archive_order":20,"of":31,"metrics":{"Diversity":"10.847","FID":"0.497","Multimodality":"1.907","R Precision Top3":"0.396"},"uses_additional_data":false},{"leaderboard":"/sota/motion-synthesis-on-motion-x","task":"Motion Synthesis","dataset":"Motion-X","model":"MDM","rank_in_archive_order":4,"of":4,"metrics":{"Diversity":"11.400","FID":"3.800","MModality":"2.530","TMR-Matching Score":"0.840","TMR-R-Precision Top3":"0.6341"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2209.14916","atlas_url":"https://app.syntology.ai/?focus=2209.14916","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2209.14916"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/guytevet/motion-diffusion-model","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1},"by_repo_kind":{},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"fe28ada12e5b4764","entry":"construct_template_variables","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"fe28ada12e5b4764"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}