Weakly Supervised Temporal Action Localization Based on Feature Enhancement

Weakly Supervised Temporal Action Localization Based on Feature Enhancement | 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 Weakly Supervised Temporal Action Localization Based on Feature Enhancement Hongying Zhang, Yi Yao This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6719748/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 21 Feb, 2026 Read the published version in Cognitive Computation → Version 1 posted You are reading this latest preprint version Abstract Weakly-supervised Temporal Action Localization (WTAL) aims to accuratelylocalize and classify action instances in untrimmed long videos using only video-level annotations. Although most existing WTAL methods leverage pre-trainedfeature extractors to obtain RGB and optical flow features—thereby reducing computational costs—this strategy suffers from two critical limitations: (1)limited temporal receptive fields, resulting in inadequate exploitation of contextual information; and (2) interference from irrelevant background content,which degrades overall performance. To address these issues, we propose aFeature-Enhanced Network (FE-Net), which comprises three key components: theLocal Feature Expansion and Enhancement Module (LF-EEM), the Cross-modalFusion Enhancement Module (CEM), and the Cross-temporal Gated FeatureFusion Module (CGFF). Specifically, LF-EEM expands the temporal receptivefield to better capture complete action instances. CEM leverages the complementary nature of auxiliary and primary modalities to suppress background noise inthe primary modality through cross-modal fusion. Furthermore, CGFF employsa cross-temporal gating mechanism during feature fusion to emphasize salientchanges across time, replacing simple concatenation. Extensive experiments conducted on two large-scale benchmark datasets, THUMOS-14 and ActivityNetv1.2, demonstrate that FE-Net significantly enhances the performance of existingWTAL methods. These results validate the effectiveness of our proposed modulesand provide new insights for advancing temporal action localization under weak supervision. Temporal Action Localization Weakly-supervised Feature-Enhanced Cross-modal Fusion Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 21 Feb, 2026 Read the published version in Cognitive Computation → 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-6719748","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":467965267,"identity":"1245ac1e-a041-4ec2-a229-ddd5461b84b7","order_by":0,"name":"Hongying Zhang","email":"","orcid":"","institution":"Civil Aviation University of China","correspondingAuthor":false,"prefix":"","firstName":"Hongying","middleName":"","lastName":"Zhang","suffix":""},{"id":467965268,"identity":"bb67b91b-eaaf-4a98-aec1-3f9d629bd7da","order_by":1,"name":"Yi Yao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA40lEQVRIie3PsWrDMBDGcYFBXUS0noixX+FAYPI4OgLeEgJZPISiocRDEroG+hIZOyoElEXZPcZvkGwdMpTMLba7ddBvvj98x1gU/UM8Pzt3Q8hkfjpeTbXqT0YgqN0vJlrt+RSvwfcnGTCtxa0ibESh2rdkwLCxLUEgGPVhi4osZ7LemO4kPXoAhLlMXdnQZ8ogXA7dCTMlIMJSWeMbCpwhzHqTAgwCHRytF7ROBiRgNLpn0kw5G5aIQK1F0GrrEzDBi95f8rp2p8fjNZMv7/f7V7XKZL3rTn4QfzuPoiiKfvUN/oRLAtgbJ1QAAAAASUVORK5CYII=","orcid":"","institution":"Civil Aviation University of China","correspondingAuthor":true,"prefix":"","firstName":"Yi","middleName":"","lastName":"Yao","suffix":""}],"badges":[],"createdAt":"2025-05-21 23:53:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6719748/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6719748/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s12559-026-10560-3","type":"published","date":"2026-02-21T15:57:51+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":103251081,"identity":"e5c42c0d-6ad2-44b9-9551-551ad9fcb257","added_by":"auto","created_at":"2026-02-23 16:03:51","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":621510,"visible":true,"origin":"","legend":"","description":"","filename":"springer3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6719748/v1_covered_6f30ae42-d73e-499b-8bff-84010efde597.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Weakly Supervised Temporal Action Localization Based on Feature Enhancement","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":true,"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":"Temporal Action Localization , Weakly-supervised, Feature-Enhanced, Cross-modal Fusion","lastPublishedDoi":"10.21203/rs.3.rs-6719748/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6719748/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWeakly-supervised Temporal Action Localization (WTAL) aims to accuratelylocalize and classify action instances in untrimmed long videos using only video-level annotations. Although most existing WTAL methods leverage pre-trainedfeature extractors to obtain RGB and optical flow features—thereby reducing computational costs—this strategy suffers from two critical limitations: (1)limited temporal receptive fields, resulting in inadequate exploitation of contextual information; and (2) interference from irrelevant background content,which degrades overall performance. To address these issues, we propose aFeature-Enhanced Network (FE-Net), which comprises three key components: theLocal Feature Expansion and Enhancement Module (LF-EEM), the Cross-modalFusion Enhancement Module (CEM), and the Cross-temporal Gated FeatureFusion Module (CGFF). Specifically, LF-EEM expands the temporal receptivefield to better capture complete action instances. CEM leverages the complementary nature of auxiliary and primary modalities to suppress background noise inthe primary modality through cross-modal fusion. Furthermore, CGFF employsa cross-temporal gating mechanism during feature fusion to emphasize salientchanges across time, replacing simple concatenation. Extensive experiments conducted on two large-scale benchmark datasets, THUMOS-14 and ActivityNetv1.2, demonstrate that FE-Net significantly enhances the performance of existingWTAL methods. These results validate the effectiveness of our proposed modulesand provide new insights for advancing temporal action localization under weak supervision.\u003c/p\u003e","manuscriptTitle":"Weakly Supervised Temporal Action Localization Based on Feature Enhancement","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-09 08:04:32","doi":"10.21203/rs.3.rs-6719748/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":"aecc957f-b823-4380-8751-dbff54ce0514","owner":[],"postedDate":"June 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-02-23T16:01:20+00:00","versionOfRecord":{"articleIdentity":"rs-6719748","link":"https://doi.org/10.1007/s12559-026-10560-3","journal":{"identity":"cognitive-computation","isVorOnly":false,"title":"Cognitive Computation"},"publishedOn":"2026-02-21 15:57:51","publishedOnDateReadable":"February 21st, 2026"},"versionCreatedAt":"2025-06-09 08:04:32","video":"","vorDoi":"10.1007/s12559-026-10560-3","vorDoiUrl":"https://doi.org/10.1007/s12559-026-10560-3","workflowStages":[]},"version":"v1","identity":"rs-6719748","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6719748","identity":"rs-6719748","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}