Swin-CATPN: A Context-Enhanced Temporal Pyramid Network with Swin Transformer for Action Recognition

Swin-CATPN: A Context-Enhanced Temporal Pyramid Network with Swin Transformer for Action Recognition | 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 Swin-CATPN: A Context-Enhanced Temporal Pyramid Network with Swin Transformer for Action Recognition Hanyou Huang, Changnan Jiang, Ziyuan Zhang, Heqing Ouyang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7069952/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 Action detection aims to identify the category and temporal boundaries (start and end times) of each action within a video sequence. It has wide applications in areas such as human-computer interaction. With the advancement of deep learning, the accuracy of action detection has significantly improved. However, challenges remain, such as the difficulty in precisely segmenting consecutive actions and the ambiguous boundaries between subjects and backgrounds in videos, which may lead to inaccurate predictions. In this work, we propose a novel Transformer-based action detection framework, termed Swin-CATPN, which integrates a Contextual Attention Memory (CAM) module and a Deformable Convolution Network (DCN) to enhance both detection precision and robustness. Extensive experiments on three benchmark datasets—Kinetics-400, Something-Something V1/V2, and EPIC-Kitchens—demonstrate that our model consistently outperforms existing state-of-the-art approaches in terms of Top-1 accuracy. Action Recognition Temporal Pyramid Network Tempo Modeling Full Text Additional Declarations No competing interests reported. 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-7069952","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":506617291,"identity":"828a30b4-37fb-467b-8bae-235013dfc46c","order_by":0,"name":"Hanyou Huang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFklEQVRIiWNgGAWjYBACfvbGhgMJBhI8/OzNB8AiEiCCB48WyZ7DBx98KLCQk+w5lsBwgBgtBjfSkg1nfKgwNriRY0CcFoYDOWbSPAYSiQ03cj5//tjGkCc5I4Hxwds2BnlzHDoYG85AtDT2vN0mcbCNoVhaIoHZcG4bg+HOBuxamBl7IFqa2XO3MQC1JM6TSGCT5m1jSDA4gF0LGzMPREsbQ87jD1At7L/xaeFhYwN630DCmIcjhwHksMTZQFuY8WmR4GEGBrKBhJwEzzEziTPnJIolex42S845J2G4AYcW+/sPgVH5p47H/njz4w8VZTZ5EseTD354U2Yjj8sWDFsTQKHIAI0e4kAC8UpHwSgYBaNgpAAAGQBdSaIWoIMAAAAASUVORK5CYII=","orcid":"","institution":"Peking University","correspondingAuthor":true,"prefix":"","firstName":"Hanyou","middleName":"","lastName":"Huang","suffix":""},{"id":506617292,"identity":"a3eb6102-c633-4f28-bd58-eca8ff581485","order_by":1,"name":"Changnan Jiang","email":"","orcid":"","institution":"Aviation Industry Corporation of China, Chinese Aeronautical Establishment","correspondingAuthor":false,"prefix":"","firstName":"Changnan","middleName":"","lastName":"Jiang","suffix":""},{"id":506617294,"identity":"8d0d9569-3738-4b43-8c64-9e403f561e52","order_by":2,"name":"Ziyuan Zhang","email":"","orcid":"","institution":"Aviation Industry Corporation of China, Chinese Aeronautical Establishment","correspondingAuthor":false,"prefix":"","firstName":"Ziyuan","middleName":"","lastName":"Zhang","suffix":""},{"id":506617295,"identity":"06bf58b5-ea6e-4a4a-bc0f-a7f0fb4f974e","order_by":3,"name":"Heqing Ouyang","email":"","orcid":"","institution":"Aviation Industry Corporation of China, Chinese Aeronautical Establishment","correspondingAuthor":false,"prefix":"","firstName":"Heqing","middleName":"","lastName":"Ouyang","suffix":""}],"badges":[],"createdAt":"2025-07-08 03:08:22","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7069952/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7069952/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96801579,"identity":"4f372261-62fe-435a-99eb-3b59c84ced6e","added_by":"auto","created_at":"2025-11-26 08:39:27","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":880347,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7069952/v1_covered_55122da7-dc89-45e9-94d4-7cc307ffab59.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Swin-CATPN: A Context-Enhanced Temporal Pyramid Network with Swin Transformer for Action Recognition","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"Action Recognition, Temporal Pyramid Network, Tempo Modeling","lastPublishedDoi":"10.21203/rs.3.rs-7069952/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7069952/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAction detection aims to identify the category and temporal boundaries (start and end times) of each action within a video sequence. It has wide applications in areas such as human-computer interaction. With the advancement of deep learning, the accuracy of action detection has significantly improved. However, challenges remain, such as the difficulty in precisely segmenting consecutive actions and the ambiguous boundaries between subjects and backgrounds in videos, which may lead to inaccurate predictions. In this work, we propose a novel Transformer-based action detection framework, termed Swin-CATPN, which integrates a Contextual Attention Memory (CAM) module and a Deformable Convolution Network (DCN) to enhance both detection precision and robustness. Extensive experiments on three benchmark datasets\u0026mdash;Kinetics-400, Something-Something V1/V2, and EPIC-Kitchens\u0026mdash;demonstrate that our model consistently outperforms existing state-of-the-art approaches in terms of Top-1 accuracy.\u003c/p\u003e","manuscriptTitle":"Swin-CATPN: A Context-Enhanced Temporal Pyramid Network with Swin Transformer for Action Recognition","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-02 07:09:58","doi":"10.21203/rs.3.rs-7069952/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":"6a1b06da-f034-439f-95da-750f7616a107","owner":[],"postedDate":"September 2nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-11-26T08:39:08+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-02 07:09:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7069952","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7069952","identity":"rs-7069952","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}