TCFNet: An End-to-End Framework for Multimodal Action Quality Assessment via Temporal Enhancement and Contrastive Fusion

TCFNet: An End-to-End Framework for Multimodal Action Quality Assessment via Temporal Enhancement and Contrastive Fusion | 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 TCFNet: An End-to-End Framework for Multimodal Action Quality Assessment via Temporal Enhancement and Contrastive Fusion Zhenxian Lin, Minghui Zhang, Chengmao Wu, Mingzhu Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7979645/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Existing Action Quality Assessment (AQA) methods have limitations, such as over reliance on single modalities, inadequate long-term temporal modeling, and modality alignment biases in multimodal models. To address these issues, we propose TCFNet, an AQA approach from the perspective of multimodal fusion. Compared to previous methods, TCFNet comprehensively integrates complementary information from multiple modalities, including RGB, optical flow, and audio. It also incorporates specialized modules to capture long-range temporal dependencies and enhance rhythmic consistency. In the single modal processing stage, we first introduce a Temporal Feature Enhancement Module (TFEM) to capture the sequential dependencies. This is followed by a three-layer pyramid network to extract multi-scale features. Then, the isomorphic multimodal fusion network receives the resulting RGB, optical flow, and audio features as 1 input. We incorporate the cross-trimodal Information Noise Contrastive Estimation loss into the loss function. This promotes feature similarity alignment and alleviates semantic and temporal discrepancies between modalities. This mechanism facilitates improved alignment of feature similarities and mitigates semantic and temporal discrepancies across modalities. Experimental results demonstrate that, compared to state-of-the-art AQA methods, our approach achieves average improvements in Spearman’s rank correlation coefficient of 4.3% on the RG dataset and 1.8% on the Fis-V dataset. Action Quality Assessment Multimodal feature Temporal Dependencies Feature alignment Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 26 Apr, 2026 Reviews received at journal 22 Apr, 2026 Reviews received at journal 20 Apr, 2026 Reviewers agreed at journal 31 Mar, 2026 Reviewers agreed at journal 31 Mar, 2026 Reviewers invited by journal 17 Dec, 2025 Editor assigned by journal 02 Dec, 2025 Submission checks completed at journal 31 Oct, 2025 First submitted to journal 29 Oct, 2025 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-7979645","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":561680058,"identity":"ac634c11-3b7d-42b5-859b-a386f923e70f","order_by":0,"name":"Zhenxian Lin","email":"","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Zhenxian","middleName":"","lastName":"Lin","suffix":""},{"id":561680059,"identity":"63b0b191-862a-4194-9b1d-fe93cb0d2971","order_by":1,"name":"Minghui Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuUlEQVRIiWNgGAWjYBACPgYeBoaEChs5NvbmA8RpYQNpeXAmzZiP51gC8VoYH7YdSpwnkaNApBb+s8ckEtgOpLcx5DAw/KjYRowt55INEnju5LYxnD3A2HPmNhFaGHsMHyRIPMttY+xLYGZsI0YLM4/BgQSDw+kgBpFa2HiAtiQcTgAyiNXCw2NskHAgzbCNhy3hIFF+4ec/Yyb585+NvPz8xwcf/KggQgsKOECi+lEwCkbBKBgFuAAAU+U3QuYE0CcAAAAASUVORK5CYII=","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":true,"prefix":"","firstName":"Minghui","middleName":"","lastName":"Zhang","suffix":""},{"id":561680060,"identity":"ea5121a1-af40-456f-91ee-3e0e8e1c1f1d","order_by":2,"name":"Chengmao Wu","email":"","orcid":"","institution":"Xi’an University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Chengmao","middleName":"","lastName":"Wu","suffix":""},{"id":561680061,"identity":"eb0fed18-881b-4812-affd-1c4e37acac5a","order_by":3,"name":"Mingzhu Zhang","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Mingzhu","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-10-29 12:08:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7979645/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7979645/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98627906,"identity":"a2f28a2d-7f9a-4bc2-994c-da889f23a297","added_by":"auto","created_at":"2025-12-19 17:10:46","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1829323,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7979645/v1/3d4144f13ee62f2bcfc66dcf.pdf"},{"id":98566595,"identity":"15e3178f-0487-4211-86f6-05a874d03794","added_by":"auto","created_at":"2025-12-19 04:45:30","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5734,"visible":true,"origin":"","legend":"","description":"","filename":"a70e35c1a4e24c90852ea4e818bcc8e3.json","url":"https://assets-eu.researchsquare.com/files/rs-7979645/v1/a2eecdee6f2c87965ca7e7f5.json"},{"id":98632031,"identity":"98606000-ca64-4cb3-b033-15341a48dded","added_by":"auto","created_at":"2025-12-19 17:20:51","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1358926,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7979645/v1_covered_7303e24b-bd9f-4360-b190-bd463dcdde44.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"TCFNet: An End-to-End Framework for Multimodal Action Quality Assessment via Temporal Enhancement and Contrastive Fusion","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"multimedia-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mmsj","sideBox":"Learn more about [Multimedia Systems](http://link.springer.com/journal/530)","snPcode":"530","submissionUrl":"https://submission.nature.com/new-submission/530/3","title":"Multimedia Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Action Quality Assessment, Multimodal feature, Temporal Dependencies, Feature alignment","lastPublishedDoi":"10.21203/rs.3.rs-7979645/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7979645/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Existing Action Quality Assessment (AQA) methods have limitations, such as over reliance on single modalities, inadequate long-term temporal modeling, and modality alignment biases in multimodal models. To address these issues, we propose TCFNet, an AQA approach from the perspective of multimodal fusion. Compared to previous methods, TCFNet comprehensively integrates complementary information from multiple modalities, including RGB, optical flow, and audio. It also incorporates specialized modules to capture long-range temporal dependencies and enhance rhythmic consistency. In the single modal processing stage, we first introduce a Temporal Feature Enhancement Module (TFEM) to capture the sequential dependencies. This is followed by a three-layer pyramid network to extract multi-scale features. Then, the isomorphic multimodal fusion network receives the resulting RGB, optical flow, and audio features as 1 input. We incorporate the cross-trimodal Information Noise Contrastive Estimation loss into the loss function. This promotes feature similarity alignment and alleviates semantic and temporal discrepancies between modalities. This mechanism facilitates improved alignment of feature similarities and mitigates semantic and temporal discrepancies across modalities. Experimental results demonstrate that, compared to state-of-the-art AQA methods, our approach achieves average improvements in Spearman’s rank correlation coefficient of 4.3% on the RG dataset and 1.8% on the Fis-V dataset.","manuscriptTitle":"TCFNet: An End-to-End Framework for Multimodal Action Quality Assessment via Temporal Enhancement and Contrastive Fusion","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-19 04:45:25","doi":"10.21203/rs.3.rs-7979645/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-26T13:14:24+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-23T02:51:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-20T14:45:40+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"197220544819625359364584146388122046070","date":"2026-04-01T02:50:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"34348151786701505230559484897245604833","date":"2026-04-01T02:43:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-17T08:22:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-03T03:58:19+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-01T02:51:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"Multimedia Systems","date":"2025-10-29T12:00:13+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"multimedia-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mmsj","sideBox":"Learn more about [Multimedia Systems](http://link.springer.com/journal/530)","snPcode":"530","submissionUrl":"https://submission.nature.com/new-submission/530/3","title":"Multimedia Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"94e613bb-7523-4e58-a70c-10f26624ce67","owner":[],"postedDate":"December 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-25T06:46:44+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-19 04:45:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7979645","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7979645","identity":"rs-7979645","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}