A Lightweight Perceptual-Guided VQVAE for High-Fidelity Image Compression

A Lightweight Perceptual-Guided VQVAE for High-Fidelity Image Compression | 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 Article A Lightweight Perceptual-Guided VQVAE for High-Fidelity Image Compression Zhisong Bie, Yunyang Kuang, Haobo Lei, Hongxia Bie This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9046607/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract This study addresses the quality-efficiency trade-off in generative models for low-bit-rate image compression by proposing HiRes-VQ, a lightweight perceptual-guided VQ-VAE framework. The framework inherits the efficient hierarchical quantization architecture of VQ-VAE-2 and innovatively introduces a super-resolution codec. By employing parallel pathways for low-frequency structure reconstruction and high-frequency detail restoration, it achieves decoupling of frequency-domain features. Additionally, a multi-scale perceptual alignment loss is adopted to guide the model in learning feature representations aligned with human visual perception. Experiments on the FFHQ-256 and ImageNet-256 datasets demonstrate that, our model significantly outperforms lightweight baseline methods across all metrics, and surpasses high-complexity models in terms of quality-efficiency balance. Physical sciences/Engineering Physical sciences/Mathematics and computing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 21 Apr, 2026 Reviews received at journal 12 Apr, 2026 Reviewers agreed at journal 12 Apr, 2026 Reviewers agreed at journal 10 Apr, 2026 Reviews received at journal 10 Apr, 2026 Reviewers agreed at journal 01 Apr, 2026 Reviewers agreed at journal 23 Mar, 2026 Reviewers invited by journal 17 Mar, 2026 Editor assigned by journal 17 Mar, 2026 Editor invited by journal 16 Mar, 2026 Submission checks completed at journal 11 Mar, 2026 First submitted to journal 11 Mar, 2026 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-9046607","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":607324882,"identity":"80f15550-2486-4ad3-9908-ecaf05b69015","order_by":0,"name":"Zhisong Bie","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYDACCcYGEJXAxsB8DCqUQLQWtjQgbUCMFrjJPGbEaZGf3dz84eOO2jw+9p5vj3l3/GHgZ88xYPi5A7cWxjkH2yRnnjlezMZzdrsx7xkDBsmeNwaMvWdwa2GWSGxj5m07ltgmkbtNmrfNgMHgRo4BM2Mbbi1sEonNn8Fa5N88A2uxJ6SFRyKxAaiyBmgLDxvEFgkCWiSADpOc2XYA6Jc0c8O5Z4x5JM48KzjYi0eL/Iz0xx8+ttXlybcffvbg7Q45Of725I0PfuLRAgWHIRQwWnlA9AGCGhgY6uBaRsEoGAWjYBRgAAA0j03GHl1NwQAAAABJRU5ErkJggg==","orcid":"","institution":"Beijing University of Posts and Telecommunications","correspondingAuthor":true,"prefix":"","firstName":"Zhisong","middleName":"","lastName":"Bie","suffix":""},{"id":607324883,"identity":"02b0bba3-4b69-46e4-92f8-961ba13993ba","order_by":1,"name":"Yunyang Kuang","email":"","orcid":"","institution":"Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Yunyang","middleName":"","lastName":"Kuang","suffix":""},{"id":607324884,"identity":"0f062563-c845-4933-990a-68f4d45eaf05","order_by":2,"name":"Haobo Lei","email":"","orcid":"","institution":"Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Haobo","middleName":"","lastName":"Lei","suffix":""},{"id":607324885,"identity":"1ac98dea-a1df-45a4-9918-a09dab03c6f0","order_by":3,"name":"Hongxia Bie","email":"","orcid":"","institution":"Beijing University of Posts and Telecommunications","correspondingAuthor":false,"prefix":"","firstName":"Hongxia","middleName":"","lastName":"Bie","suffix":""}],"badges":[],"createdAt":"2026-03-06 06:09:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9046607/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9046607/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105034547,"identity":"76417f50-724e-43a4-a0c5-02b7f980fcbf","added_by":"auto","created_at":"2026-03-20 07:23:33","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3527232,"visible":true,"origin":"","legend":"","description":"","filename":"HiResVQ.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9046607/v1_covered_93a7dde9-16a4-4aba-ae66-9e7fc5f5bf37.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Lightweight Perceptual-Guided VQVAE for High-Fidelity Image Compression","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9046607/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9046607/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"This study addresses the quality-efficiency trade-off in generative models for low-bit-rate image compression by proposing HiRes-VQ, a lightweight perceptual-guided VQ-VAE framework. The framework inherits the efficient hierarchical quantization architecture of VQ-VAE-2 and innovatively introduces a super-resolution codec. By employing parallel pathways for low-frequency structure reconstruction and high-frequency detail restoration, it achieves decoupling of frequency-domain features. Additionally, a multi-scale perceptual alignment loss is adopted to guide the model in learning feature representations aligned with human visual perception. Experiments on the FFHQ-256 and ImageNet-256 datasets demonstrate that, our model significantly outperforms lightweight baseline methods across all metrics, and surpasses high-complexity models in terms of quality-efficiency balance. ","manuscriptTitle":"A Lightweight Perceptual-Guided VQVAE for High-Fidelity Image Compression","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-18 10:45:06","doi":"10.21203/rs.3.rs-9046607/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-21T07:03:55+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-12T14:24:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"202024677268096149944163362311977479049","date":"2026-04-12T13:14:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"324711630582530547233275946909496007824","date":"2026-04-10T15:28:52+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-10T06:24:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"182558520381708405574968033837842673605","date":"2026-04-01T06:20:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"81946177369509619876837062173529812194","date":"2026-03-23T05:46:25+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-17T05:21:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-17T05:05:42+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-16T09:29:35+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-11T20:58:59+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-03-11T13:47:28+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c54bb02a-f21a-4cbe-8efd-bc1945decb3e","owner":[],"postedDate":"March 18th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64623662,"name":"Physical sciences/Engineering"},{"id":64623663,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-14T10:24:16+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-18 10:45:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9046607","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9046607","identity":"rs-9046607","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}