UASR: An Unified-Attention Mixer Network for Efficient Image Super-Resolution

UASR: An Unified-Attention Mixer Network for Efficient Image Super-Resolution | 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 UASR: An Unified-Attention Mixer Network for Efficient Image Super-Resolution Yuqiao Huang, Wei Zhou, Zhaowen Huang, Tao Su, Dihu Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6672339/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 Recent works in single-image super-resolution (SISR) have brought notable improvements to the field. Transformer-based methods enhance reconstruction quality by capturing long-range dependencies. However, the quadratic computational complexity of multi-head self-attention (MHSA) introduces efficiency bottlenecks in HR image processing, and insufficient local feature extraction limits the recovery of fine texture details and edge sharpness. In contrast, convolutional neural network (CNN)-based methods suffer from limited receptive fields, leading to inadequate high-frequency detail recovery and blurring artifacts. Generally, Transformer-based and CNN-based methods fail to simultaneously address the challenges of computational efficiency, global dependency modeling, and local feature extraction. To integrate the strengths of both paradigms, we propose Unified-Attention Super-Resolution(UASR) network, a lightweight architecture based on the Convolutional Transformer(ConvFormer) layer. Specifically, UASR replaces MHSA with the Unified-Attention Mixer (UA-M) that efficiently captures global dependencies at a low computational cost. Additionally, the Reparameterized Edge-Extraction FeedForward Network (REFN) supplements UA-M by focusing on extracting texture and edge features. Furthermore, we introduce a Spectral Unified-Attention Block (SUAB) that extends the capabilities of UA-M into the frequency domain, thus improving detail reconstruction and accelerating the computation process. Compared to current CNN-based and Transformer-based SISR models, experimental results demonstrate that our method strikes an effective balance between accuracy and efficiency, enhancing texture fidelity and super-resolution performance. Single Image Super-Resolution Efficient network Convolution Transformer Reparameterization Attention mechanism 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-6672339","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":475760337,"identity":"a365c564-112f-4223-a7f3-814e31277637","order_by":0,"name":"Yuqiao Huang","email":"","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Yuqiao","middleName":"","lastName":"Huang","suffix":""},{"id":475760338,"identity":"dc13024c-e3c4-4873-91de-d54a8a07cba1","order_by":1,"name":"Wei Zhou","email":"","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Zhou","suffix":""},{"id":475760339,"identity":"a1812258-50a7-44e5-abcd-55cfd9ea9ace","order_by":2,"name":"Zhaowen Huang","email":"","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Zhaowen","middleName":"","lastName":"Huang","suffix":""},{"id":475760340,"identity":"a4924085-a952-40ea-a4b1-763ba47c82ca","order_by":3,"name":"Tao Su","email":"","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Tao","middleName":"","lastName":"Su","suffix":""},{"id":475760341,"identity":"97ae746b-7c4b-473e-978c-d34de155778f","order_by":4,"name":"Dihu Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYLAC3gYbBgZm5gYY3wC/cjawljSgFkbStBwGksRqkZ/f/vDD2x3no/nbGRsYf7bVJTawN2+TYKi5g1OLwTEeY8m5Z27nzjjM2MDM23Y4sYHnWJkEw7FnuLWw8bABVd7ObQBpYWw7kNggkWMmwQh2Kg6HtbE/A2o5lzv/MMxh8m/wa2E4xmAG1HIgdwNQCwNvGzPQFh78WgyO5QD90pacuxGo5TDPucPGbTxpxRYJx/A4rPk4MMTa7HLnnT988OGPsjrZfvbDG298qMHjMGRwgJENElEMCcRpAIE/xCsdBaNgFIyCkQMA8oxVz6cta2QAAAAASUVORK5CYII=","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":true,"prefix":"","firstName":"Dihu","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2025-05-15 11:53:19","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6672339/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6672339/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88839289,"identity":"c5144fc2-609b-4ca7-8b2a-26d16c23a630","added_by":"auto","created_at":"2025-08-12 02:01:46","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3018055,"visible":true,"origin":"","legend":"","description":"","filename":"UASRFORMS.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6672339/v1_covered_efbdfd5e-aad4-41fc-ad2d-3c27ce50a260.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"UASR: An Unified-Attention Mixer Network for Efficient Image Super-Resolution","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":"Single Image Super-Resolution, Efficient network, Convolution, Transformer, Reparameterization, Attention, mechanism","lastPublishedDoi":"10.21203/rs.3.rs-6672339/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6672339/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRecent works in single-image super-resolution (SISR) have brought notable improvements to the field. Transformer-based methods enhance reconstruction quality by capturing long-range dependencies. However, the quadratic computational complexity of multi-head self-attention (MHSA) introduces efficiency bottlenecks in HR image processing, and insufficient local feature extraction limits the recovery of fine texture details and edge sharpness. In contrast, convolutional neural network (CNN)-based methods suffer from limited receptive fields, leading to inadequate high-frequency detail recovery and blurring artifacts. Generally, Transformer-based and CNN-based methods fail to simultaneously address the challenges of computational efficiency, global dependency modeling, and local feature extraction. To integrate the strengths of both paradigms, we propose Unified-Attention Super-Resolution(UASR) network, a lightweight architecture based on the Convolutional Transformer(ConvFormer) layer. Specifically, UASR replaces MHSA with the Unified-Attention Mixer (UA-M) that efficiently captures global dependencies at a low computational cost. Additionally, the Reparameterized Edge-Extraction FeedForward Network (REFN) supplements UA-M by focusing on extracting texture and edge features. Furthermore, we introduce a Spectral Unified-Attention Block (SUAB) that extends the capabilities of UA-M into the frequency domain, thus improving detail reconstruction and accelerating the computation process. Compared to current CNN-based and Transformer-based SISR models, experimental results demonstrate that our method strikes an effective balance between accuracy and efficiency, enhancing texture fidelity and super-resolution performance.\u003c/p\u003e","manuscriptTitle":"UASR: An Unified-Attention Mixer Network for Efficient Image Super-Resolution","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-25 02:49:35","doi":"10.21203/rs.3.rs-6672339/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":"4be6e5d5-e2ee-4908-b7b4-e86faa07f10c","owner":[],"postedDate":"June 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-12T01:53:36+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-25 02:49:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6672339","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6672339","identity":"rs-6672339","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}