Enhancing High-Resolution Facial Ageing with Dual Attention Wavelet GAN

Enhancing High-Resolution Facial Ageing with Dual Attention Wavelet GAN | 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 Enhancing High-Resolution Facial Ageing with Dual Attention Wavelet GAN Youran Zhi, Baole Wang, Ao Guo, Feng Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8219617/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract Generating realistic facial ageing using generative adversarial networks (GANs) has seen significant progress, yet challenges such as unnatural attribute variations, image distortions, and irrelevant content modifications persist. In this paper, we introduce a dual attention mechanism and a wavelet-based discriminator within a GAN framework to capture facial attribute texture information, enhancing the ageing effect. Our approach efficiently handles both low-resolution and high-quality facial images, preserving fine details and reducing artifacts. Through extensive experiments, the effectiveness of the dual attention mechanism and wavelet-based discriminator in enhancing model performance is demonstrated. Qualitative and quantitative analyses reveal that our model achieves facial verification rates of 99.84%, 99.23%, and 98.74% for age groups 31–40, 41–50, and 51–60, respectively, with minimal discrepancies between synthesized and target age faces. Compared to existing research, our model exhibits superior performance, effectively addressing challenges in facial ageing synthesis. Face aging Generative adversarial network attention mechanism wavelet transform high resolution Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 22 Apr, 2026 Reviewers agreed at journal 19 Apr, 2026 Reviewers invited by journal 03 Mar, 2026 Editor assigned by journal 29 Nov, 2025 Submission checks completed at journal 29 Nov, 2025 First submitted to journal 27 Nov, 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-8219617","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":600515236,"identity":"52963109-a86d-4b05-8010-0d4f744cff3a","order_by":0,"name":"Youran Zhi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYBACefmDDQc+VLDx8LM3EKnFcAZz48EZZ/hkJHsOEGvNDfbmw7xtcjYGNxKI1ME4u7HhAA+bGQ/DzccbbzDU2EQT1MIuA/SLBE8aD+PstGILhmNpuQ0EbWlIbDhgIHGMh1k6x0yCseEwYS0MB4BaEgz+87BJniFWyw2glgMJbDw8EjxEajHsOdgA9A0bUAfQLwnE+EWevf3x57//2Oztjx/eeONDjQ0RDkMCBhIJpCiHaCFVxygYBaNgFIwMAACFxEKyUNBg5QAAAABJRU5ErkJggg==","orcid":"","institution":"Nanjing Institute of Technology","correspondingAuthor":true,"prefix":"","firstName":"Youran","middleName":"","lastName":"Zhi","suffix":""},{"id":600515237,"identity":"7ee5ad75-1097-40b8-83b0-031d19c71a91","order_by":1,"name":"Baole Wang","email":"","orcid":"","institution":"Nanjing Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Baole","middleName":"","lastName":"Wang","suffix":""},{"id":600515238,"identity":"532e8228-af96-4af8-b829-8ea0375388a1","order_by":2,"name":"Ao Guo","email":"","orcid":"","institution":"Nanjing Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Ao","middleName":"","lastName":"Guo","suffix":""},{"id":600515239,"identity":"8f0ce866-5f6a-4628-9aa6-ffbf13fdb848","order_by":3,"name":"Feng Zhang","email":"","orcid":"","institution":"Nanjing Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Feng","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-11-27 08:23:46","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8219617/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8219617/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104402239,"identity":"ee844ba1-61e9-430b-8c26-c8bc27ce5b9a","added_by":"auto","created_at":"2026-03-11 12:14:45","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":820878,"visible":true,"origin":"","legend":"","description":"","filename":"faceaging.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8219617/v1_covered_67f87599-e63f-4f6a-b2a9-d46f1a1b4413.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Enhancing High-Resolution Facial Ageing with Dual Attention Wavelet GAN","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":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Face aging, Generative adversarial network, attention mechanism, wavelet transform, high resolution","lastPublishedDoi":"10.21203/rs.3.rs-8219617/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8219617/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGenerating realistic facial ageing using generative adversarial networks (GANs) has seen significant progress, yet challenges such as unnatural attribute variations, image distortions, and irrelevant content modifications persist. In this paper, we introduce a dual attention mechanism and a wavelet-based discriminator within a GAN framework to capture facial attribute texture information, enhancing the ageing effect. Our approach efficiently handles both low-resolution and high-quality facial images, preserving fine details and reducing artifacts. Through extensive experiments, the effectiveness of the dual attention mechanism and wavelet-based discriminator in enhancing model performance is demonstrated. Qualitative and quantitative analyses reveal that our model achieves facial verification rates of 99.84%, 99.23%, and 98.74% for age groups 31\u0026ndash;40, 41\u0026ndash;50, and 51\u0026ndash;60, respectively, with minimal discrepancies between synthesized and target age faces. Compared to existing research, our model exhibits superior performance, effectively addressing challenges in facial ageing synthesis.\u003c/p\u003e","manuscriptTitle":"Enhancing High-Resolution Facial Ageing with Dual Attention Wavelet GAN","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-06 04:27:30","doi":"10.21203/rs.3.rs-8219617/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"70141687803350808077471703513119888788","date":"2026-04-22T04:20:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"220530703596415022013589309066474049101","date":"2026-04-19T05:00:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-03T17:28:11+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-29T10:43:17+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-29T10:42:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Signal, Image and Video Processing","date":"2025-11-27T08:13:51+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"89eb0caa-be1f-442e-98e2-d62961459ba0","owner":[],"postedDate":"March 6th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-06T04:27:30+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-06 04:27:30","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8219617","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8219617","identity":"rs-8219617","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}