HQ-Font: Few-shot Font Generation via Transferring Hierarchical Quantization Styles

HQ-Font: Few-shot Font Generation via Transferring Hierarchical Quantization Styles | 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 HQ-Font: Few-shot Font Generation via Transferring Hierarchical Quantization Styles Anna Zhu, Wei Pan, Guan Li, Hongyi Cai, Kenji Brian This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7876044/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 26 Apr, 2026 Read the published version in International Journal on Document Analysis and Recognition (IJDAR) → Version 1 posted 10 You are reading this latest preprint version Abstract Utilizing artificial intelligence for few-shot font generation (FFG) has become a trend in designing fonts for glyph-rich scripts. Most existing FFG approaches either globally disentangle the content and style of reference glyphs or decompose glyphs into strokes or radicals, then transfer these styles component-wise. However, they may fail to distinguish fine-grained local details or require predefined decomposition rules or special infeasible training strategies. This paper introduces a Hierarchical Quantization-based FFG approach (HQ-Font) by aggregating different-grained styles. It adopts a vector quantization strategy for glyph representation through unsupervised learning, enabling the contrasting and learning of discrete latent representations from low to high-level glyph feature spaces simultaneously without manual definition. A cross-attention mechanism is employed to transfer different granular styles of reference glyphs onto the discrete latent codes through contrastive learning, generating a complete set of content-agnostic style representations for different scripts. To this end, a font generation decoder performs hierarchical font synthesis, gradually mapping the corresponding global semantic and local stroke stylized codes from a given input to the final font image output. The number of glyph references as input can vary without the need for fine-tuning during testing, making HQ-Font more flexible. The experimental results demonstrate the effectiveness and generalizability of HQ-Font across different linguistic scripts and also show its superiority when compared with other state-of-the-art FFG methods. Font generation Few-shot style transfer self-supervision vector quantization Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 26 Apr, 2026 Read the published version in International Journal on Document Analysis and Recognition (IJDAR) → Version 1 posted Editorial decision: Revision requested 16 Jan, 2026 Reviews received at journal 02 Jan, 2026 Reviews received at journal 29 Dec, 2025 Reviewers agreed at journal 18 Dec, 2025 Reviewers agreed at journal 04 Dec, 2025 Reviewers agreed at journal 02 Dec, 2025 Reviewers invited by journal 01 Dec, 2025 Editor assigned by journal 17 Oct, 2025 Submission checks completed at journal 17 Oct, 2025 First submitted to journal 16 Oct, 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-7876044","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":553902369,"identity":"397dd443-43b8-4666-a243-765e29a332fb","order_by":0,"name":"Anna Zhu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzklEQVRIiWNgGAWjYLACxgYGOQYJMJOZeC3GQC1AihQtiQ1Ea+Gf3Xvw488dh9M33O4xf8BQYZ3YwH72AF4tEnfOJUtInjmcu+HOGcMGhjPpiQ08eQn4rbmRYyBh2HY4d9uNHMMGxrbDQBfyGODVIX8jx/hHYtvhdDOwln9EaDG4kWMmcbDtcAJESwMRWgyBWiwb29IN999IK5yRcCzduI0nB78WOaDDbv5ss5aXnJG84cOHGmvZfvYz+LWgggQgZiNB/SgYBaNgFIwCHAAALPJJyGxIc1IAAAAASUVORK5CYII=","orcid":"","institution":"Wuhan University of Technology","correspondingAuthor":true,"prefix":"","firstName":"Anna","middleName":"","lastName":"Zhu","suffix":""},{"id":553902370,"identity":"15fa6738-5e41-4d83-9af7-82668f9932e7","order_by":1,"name":"Wei Pan","email":"","orcid":"","institution":"South China University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Pan","suffix":""},{"id":553902371,"identity":"b31862c9-dc0c-473e-ba58-d8abddc78239","order_by":2,"name":"Guan Li","email":"","orcid":"","institution":"Wuhan University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Guan","middleName":"","lastName":"Li","suffix":""},{"id":553902372,"identity":"ef02ca2c-b1e7-4124-8508-c07733a797d9","order_by":3,"name":"Hongyi Cai","email":"","orcid":"","institution":"Wuhan University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Hongyi","middleName":"","lastName":"Cai","suffix":""},{"id":553902373,"identity":"9a08c327-655a-4d38-adbc-972f1cddfcd6","order_by":4,"name":"Kenji Brian","email":"","orcid":"","institution":"Kyushu University","correspondingAuthor":false,"prefix":"","firstName":"Kenji","middleName":"","lastName":"Brian","suffix":""}],"badges":[],"createdAt":"2025-10-16 10:08:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7876044/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7876044/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10032-026-00580-y","type":"published","date":"2026-04-26T15:59:10+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":97309226,"identity":"60d64d8c-097c-4162-895b-16abd9b4cf8e","added_by":"auto","created_at":"2025-12-03 04:14:45","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6595,"visible":true,"origin":"","legend":"","description":"","filename":"2f46a29c897a4f15be95ea0d34093b40.json","url":"https://assets-eu.researchsquare.com/files/rs-7876044/v1/a3509bcbeeacb9739ec70e8a.json"},{"id":107928025,"identity":"252a83e4-48a2-426f-8663-438a82440f5b","added_by":"auto","created_at":"2026-04-27 16:06:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7386424,"visible":true,"origin":"","legend":"","description":"","filename":"IJDARHQFont.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7876044/v1_covered_6477e682-e0f7-49c2-b6ba-16f713f83a4c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"HQ-Font: Few-shot Font Generation via Transferring Hierarchical Quantization Styles","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":"international-journal-on-document-analysis-and-recognition-ijdar","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijda","sideBox":"Learn more about [International Journal on Document Analysis and Recognition (IJDAR)](http://link.springer.com/journal/10032)","snPcode":"10032","submissionUrl":"https://submission.nature.com/new-submission/10032/3","title":"International Journal on Document Analysis and Recognition (IJDAR)","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Font generation, Few-shot, style transfer, self-supervision, vector quantization","lastPublishedDoi":"10.21203/rs.3.rs-7876044/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7876044/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Utilizing artificial intelligence for few-shot font generation (FFG) has become a trend in designing fonts for glyph-rich scripts. Most existing FFG approaches either globally disentangle the content and style of reference glyphs or decompose glyphs into strokes or radicals, then transfer these styles component-wise. However, they may fail to distinguish fine-grained local details or require predefined decomposition rules or special infeasible training strategies. This paper introduces a Hierarchical Quantization-based FFG approach (HQ-Font) by aggregating different-grained styles. It adopts a vector quantization strategy for glyph representation through unsupervised learning, enabling the contrasting and learning of discrete latent representations from low to high-level glyph feature spaces simultaneously without manual definition. A cross-attention mechanism is employed to transfer different granular styles of reference glyphs onto the discrete latent codes through contrastive learning, generating a complete set of content-agnostic style representations for different scripts. To this end, a font generation decoder performs hierarchical font synthesis, gradually mapping the corresponding global semantic and local stroke stylized codes from a given input to the final font image output. The number of glyph references as input can vary without the need for fine-tuning during testing, making HQ-Font more flexible. The experimental results demonstrate the effectiveness and generalizability of HQ-Font across different linguistic scripts and also show its superiority when compared with other state-of-the-art FFG methods.","manuscriptTitle":"HQ-Font: Few-shot Font Generation via Transferring Hierarchical Quantization Styles","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-03 04:14:41","doi":"10.21203/rs.3.rs-7876044/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-16T12:22:03+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-02T23:28:30+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-30T02:50:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"245532957210967939854117196490185968070","date":"2025-12-18T09:31:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"77674057782020702974751757490525007293","date":"2025-12-05T03:55:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"228107897139266824642057811016877396101","date":"2025-12-02T08:04:13+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-01T12:33:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-18T03:36:25+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-17T14:54:27+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal on Document Analysis and Recognition (IJDAR)","date":"2025-10-16T09:52:55+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-on-document-analysis-and-recognition-ijdar","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ijda","sideBox":"Learn more about [International Journal on Document Analysis and Recognition (IJDAR)](http://link.springer.com/journal/10032)","snPcode":"10032","submissionUrl":"https://submission.nature.com/new-submission/10032/3","title":"International Journal on Document Analysis and Recognition (IJDAR)","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"723dd40e-69bf-4310-b82f-f9a12778fa35","owner":[],"postedDate":"December 3rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T16:03:57+00:00","versionOfRecord":{"articleIdentity":"rs-7876044","link":"https://doi.org/10.1007/s10032-026-00580-y","journal":{"identity":"international-journal-on-document-analysis-and-recognition-ijdar","isVorOnly":false,"title":"International Journal on Document Analysis and Recognition (IJDAR)"},"publishedOn":"2026-04-26 15:59:10","publishedOnDateReadable":"April 26th, 2026"},"versionCreatedAt":"2025-12-03 04:14:41","video":"","vorDoi":"10.1007/s10032-026-00580-y","vorDoiUrl":"https://doi.org/10.1007/s10032-026-00580-y","workflowStages":[]},"version":"v1","identity":"rs-7876044","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7876044","identity":"rs-7876044","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}