Vehicle Re-Identification via Multi-Scale Feature Learningand Dual Attention Fusion.

Vehicle Re-Identification via Multi-Scale Feature Learningand Dual Attention 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 Vehicle Re-Identification via Multi-Scale Feature Learningand Dual Attention Fusion. Haifeng Sang, Bochi Zhu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6308474/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 In the continuous evolution of intelligent transportation systems, vehicle re-identification technology faces numerous technical challenges, including variations in perspective and equipment resolution. These factors lead to significant intra-class discrepancies in the performance of identical vehicles under varying conditions, as well as inter-class confusion among vehicles with similar appearances. To address these challenges, we integrate vehicle color and type attribute information, enhancing the model’s ability to capture semantic features and improve its discriminative performance. Additionally, we propose a wavelet feature enhancement module that employs wavelet transform to decompose images at multiple scales, effectively capturing fine-grained features such as edges and textures. This enables the model to better represent intricate visual details. Finally, we introduce a dual attention mechanism that combines global and local features, strengthening contextual understanding through interactive feature modeling. Experimental results demonstrate the effectiveness of our approach, achieving a Rank-1 accuracy of 96.8 percent on the VeRi-776 dataset and 84.9 percent on the VehicleID dataset, outperforming existing methods and highlighting the efficacy of our proposed framework. Vehicle re-identification Wavelet transform Attribute aggregation Swin transformer 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-6308474","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":444776716,"identity":"029d035f-045d-4e3c-be92-3b85404de245","order_by":0,"name":"Haifeng Sang","email":"","orcid":"","institution":"Shenyang University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Haifeng","middleName":"","lastName":"Sang","suffix":""},{"id":444776718,"identity":"214f072d-d466-4642-acd6-966e06f55fbd","order_by":1,"name":"Bochi Zhu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAApElEQVRIiWNgGAWjYDACCSBmbLDh4edvIE1LmozkjAOkaTlsY9CQQKQO+dk9ZtK8O87zGDAcYPzwMYcILYxzzgC1nLnNY87cwCw5cxsRWpglcoBa2m7zWDYcYGPmJUYLG0TLOR6DAwlEauGBaDlAghYJibRiy7lnknkkZxxsJs4v8jOSN954u8POnp+/+eCHj8RoQQKMDaSpHwWjYBSMglGAGwAA8Z0v1UPTrUgAAAAASUVORK5CYII=","orcid":"","institution":"Shenyang University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Bochi","middleName":"","lastName":"Zhu","suffix":""}],"badges":[],"createdAt":"2025-03-26 04:23:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6308474/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6308474/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82120512,"identity":"13b168d6-b84c-4adf-9f5a-262daf6b0ade","added_by":"auto","created_at":"2025-05-07 03:16:35","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":705932,"visible":true,"origin":"","legend":"","description":"","filename":"VehicleReIdentificationBasedonWaveletFeatureEnhancementandDualAttentionFusion..pdf","url":"https://assets-eu.researchsquare.com/files/rs-6308474/v1_covered_bc93943c-403e-403e-9ee7-7afaa11436e0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Vehicle Re-Identification via Multi-Scale Feature Learningand Dual Attention Fusion.","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":"Vehicle re-identification, Wavelet transform, Attribute aggregation, Swin transformer","lastPublishedDoi":"10.21203/rs.3.rs-6308474/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6308474/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn the continuous evolution of intelligent transportation systems, vehicle re-identification technology faces numerous technical challenges, including variations in perspective and equipment resolution. These factors lead to significant intra-class discrepancies in the performance of identical vehicles under varying conditions, as well as inter-class confusion among vehicles with similar appearances. To address these challenges, we integrate vehicle color and type attribute information, enhancing the model\u0026rsquo;s ability to capture semantic features and improve its discriminative performance. Additionally, we propose a wavelet feature enhancement module that employs wavelet transform to decompose images at multiple scales, effectively capturing fine-grained features such as edges and textures. This enables the model to better represent intricate visual details. Finally, we introduce a dual attention mechanism that combines global and local features, strengthening contextual understanding through interactive feature modeling. Experimental results demonstrate the effectiveness of our approach, achieving a Rank-1 accuracy of 96.8 percent on the VeRi-776 dataset and 84.9 percent on the VehicleID dataset, outperforming existing methods and highlighting the efficacy of our proposed framework.\u003c/p\u003e","manuscriptTitle":"Vehicle Re-Identification via Multi-Scale Feature Learningand Dual Attention Fusion.","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-21 08:21:57","doi":"10.21203/rs.3.rs-6308474/v1","editorialEvents":[{"type":"communityComments","content":1}],"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":"4c1596fa-56e6-4b58-8501-2dcab98c2270","owner":[],"postedDate":"April 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-05-07T03:08:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-21 08:21:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6308474","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6308474","identity":"rs-6308474","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}