A Unified Framework for Human Motion Generation with Multimodal Inputs

A Unified Framework for Human Motion Generation with Multimodal Inputs | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A Unified Framework for Human Motion Generation with Multimodal Inputs Nathan J. Blake, Isabella M. Cooper, Ryan A. Mitchell, Chloe S. Turner This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7467386/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract To enable generalized human motion generation, this paper proposes a unified generation framework, UniMotion, which supports multimodal inputs including text, image and audio. The method uses a unified prompt encoder to map different inputs into a shared cross-modal semantic space. It adopts a two-stage motion decoder to gradually generate fine-grained skeleton sequences. A multimodal alignment loss function is introduced to strengthen consistency modeling across different prompts. In semantic generalization evaluation and prompt consistency tests, UniMotion outperforms baseline methods by 7.3% and 8.9%, respectively. In random multimodal prompt switching tests, it maintains 92.4% motion stability and logical consistency, demonstrating good practicality and scalability. This study expands the application scope of multimodal generative models in human motion modeling. Artificial Intelligence and Machine Learning Theoretical Computer Science Human motion synthesis Multimodal prompts Unified modeling Text-image-audio fusion Cross-modal generation Training without motion capture Large model-driven approach Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7467386","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":506126510,"identity":"0b05ef5e-d715-4bea-9806-a144c0bc4b83","order_by":0,"name":"Nathan J. Blake","email":"","orcid":"","institution":"Department of Computer Sciences, University of Wisconsin–Madison, Madison, USA","correspondingAuthor":false,"prefix":"","firstName":"Nathan","middleName":"J.","lastName":"Blake","suffix":""},{"id":506126511,"identity":"e92c5a67-f0de-4fb7-8160-7401b6760eab","order_by":1,"name":"Isabella M. Cooper","email":"","orcid":"","institution":"Department of Computer Sciences, University of Wisconsin–Madison, Madison, USA","correspondingAuthor":false,"prefix":"","firstName":"Isabella","middleName":"M.","lastName":"Cooper","suffix":""},{"id":506126512,"identity":"0b16a54b-4d30-4346-952e-087498533556","order_by":2,"name":"Ryan A. Mitchell","email":"","orcid":"","institution":"Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Urbana, USA","correspondingAuthor":false,"prefix":"","firstName":"Ryan","middleName":"A.","lastName":"Mitchell","suffix":""},{"id":506126513,"identity":"0a55cf64-f466-4362-845e-dc6c94658dde","order_by":3,"name":"Chloe S. Turner","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIie3PMQrCMBSA4dclLoGsEaReIeIggtSrpBR0UfAImXQRuvYYHuGVgFPsnFEXJwfHDgVtBdEp1U0w/xAeIR+PAPh8P5kJEBDCxyzfTheBhgy/IfRBYvW6aSGj4IBYmsk8TZHwYxUB6yyEk4xVIfOtnS0zKwmP1wl0txc3EWgE0qte7jgQGysEYVu2NCSvrre5YEisrBCmnxBNLUoBsiak3sJbSSF1zySDzMabsv4L5ea8chN7yE+XfdRnqdaDsopCtkl2TgIcn1OgmpO6nzcx1f7G5/P5/rw7gWdTYqjGjRcAAAAASUVORK5CYII=","orcid":"","institution":"Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Urbana, USA","correspondingAuthor":true,"prefix":"","firstName":"Chloe","middleName":"S.","lastName":"Turner","suffix":""}],"badges":[],"createdAt":"2025-08-27 03:25:21","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-7467386/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7467386/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90061566,"identity":"ddd6b2ff-d5f3-46f2-bd40-c4dcd18c5a7e","added_by":"auto","created_at":"2025-08-28 03:38:17","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":616997,"visible":true,"origin":"","legend":"","description":"","filename":"115AUnifiedFrameworkforHumanMotionGenerationwithMultimodalInputs.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7467386/v1_covered_5e551e64-c4c0-43f1-8404-34e5cf6dcbd9.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eA Unified Framework for Human Motion Generation with Multimodal Inputs\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Human motion synthesis, Multimodal prompts, Unified modeling, Text-image-audio fusion, Cross-modal generation, Training without motion capture, Large model-driven approach","lastPublishedDoi":"10.21203/rs.3.rs-7467386/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7467386/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eTo enable generalized human motion generation, this paper proposes a unified generation framework, UniMotion, which supports multimodal inputs including text, image and audio. The method uses a unified prompt encoder to map different inputs into a shared cross-modal semantic space. It adopts a two-stage motion decoder to gradually generate fine-grained skeleton sequences. A multimodal alignment loss function is introduced to strengthen consistency modeling across different prompts. In semantic generalization evaluation and prompt consistency tests, UniMotion outperforms baseline methods by 7.3% and 8.9%, respectively. In random multimodal prompt switching tests, it maintains 92.4% motion stability and logical consistency, demonstrating good practicality and scalability. This study expands the application scope of multimodal generative models in human motion modeling.\u003c/p\u003e","manuscriptTitle":"A Unified Framework for Human Motion Generation with Multimodal Inputs","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-28 03:30:09","doi":"10.21203/rs.3.rs-7467386/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e7349d65-e0eb-4301-a6c0-ef4e5dda3594","owner":[],"postedDate":"August 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":53759775,"name":"Artificial Intelligence and Machine Learning"},{"id":53759776,"name":"Theoretical Computer Science"}],"tags":[],"updatedAt":"2025-08-28T03:30:09+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-28 03:30:09","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7467386","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7467386","identity":"rs-7467386","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}