Geo-TCAM: A Thangka Captioning Method Integrating Topic Modeling with Geometry- Guided Spatial Attention

Geo-TCAM: A Thangka Captioning Method Integrating Topic Modeling with Geometry- Guided Spatial Attention | 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 Geo-TCAM: A Thangka Captioning Method Integrating Topic Modeling with Geometry- Guided Spatial Attention Ping Zhong, Wenjin Hu, Yinqiu Zhao, Fujun Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7108021/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 07 Feb, 2026 Read the published version in npj Heritage Science → Version 1 posted 17 You are reading this latest preprint version Abstract Thangka image captioning, an essential task in cultural heritage preservation, faces challenges due to the complexity of Thangka imagery and the depth of their semantic content. Current deep learning-based methods struggle with extracting detailed features and accurately understanding the semantics of Thangka images, often leading to incomplete or incorrect captions of key elements such as the main deity and the background. To address these challenges, this paper introduces a novel Thangka captioning model, integrating topic modeling and geometry-guided spatial attention (Geo-TCAM). The model employs a multi-level feature integration strategy to enhance feature extraction, including gestures and objects. By incorporating Latent Dirichlet Allocation (LDA) topic weights and visual features (TIF), it leverages external domain knowledge for better semantic understanding. The Geo-TCAM's geometry-guided facial spatial attention module (GFSA) improves spatial layout recognition. Experimental results demonstrate significant improvements in captioning performance, with BLEU-1, BLEU-4, METEOR, and CIDEr scores increasing by 11.9%, 17.1%, 9.7%, and 119.5%, respectively, compared to baseline models. On the COCO public dataset, the Geo-TCAM model also demonstrates outstanding performance, comparable to that of other state-of-the-art models. This study employs the Geo-TCAM model to accurately generate image captions for Thangka images, facilitating the digital preservation and dissemination of cultural heritage. Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 07 Feb, 2026 Read the published version in npj Heritage Science → Version 1 posted Editorial decision: Revision requested 12 Aug, 2025 Reviews received at journal 12 Aug, 2025 Reviews received at journal 11 Aug, 2025 Reviews received at journal 07 Aug, 2025 Reviewers agreed at journal 28 Jul, 2025 Reviewers agreed at journal 27 Jul, 2025 Reviewers agreed at journal 25 Jul, 2025 Reviewers agreed at journal 24 Jul, 2025 Reviews received at journal 24 Jul, 2025 Reviewers agreed at journal 23 Jul, 2025 Reviewers agreed at journal 23 Jul, 2025 Reviewers agreed at journal 23 Jul, 2025 Reviewers agreed at journal 23 Jul, 2025 Reviewers invited by journal 23 Jul, 2025 Editor assigned by journal 17 Jul, 2025 Submission checks completed at journal 17 Jul, 2025 First submitted to journal 12 Jul, 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-7108021","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":490304694,"identity":"6f7cf9cc-199c-45be-ade4-85122166514f","order_by":0,"name":"Ping Zhong","email":"","orcid":"","institution":"Chinese National Information Technology Research Institute, Northwest Minzu University","correspondingAuthor":false,"prefix":"","firstName":"Ping","middleName":"","lastName":"Zhong","suffix":""},{"id":490304699,"identity":"77e6d884-e707-4d0c-b1e1-b928784eb66f","order_by":1,"name":"Wenjin Hu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1UlEQVRIiWNgGAWjYBACfoaDjY9//LDh4WdvPkCcFsnGw4eNGXvS5CR7jiUQp8Xg8LE0aQa2Q8YGN3wMiHTZsTMG0gU8BxJnzuD5eOMNg52cbgMBHYw9ZwyMZ1jcSeyX7t1sOYch2djsAAEtzBJnDBJ4eJ4lzpxzdps0D8OBxG2EtLDJvzE4wMN2OHHDjZxnxGnhYTiW2AzUAvR+DhtxWiQYDh9mnAkJZGPLOQZE+MX+wMH2Hx8gUfnwxpsKOzmCWlCt5CE2apC0kKpjFIyCUTAKRgQAANVtSq5QELJNAAAAAElFTkSuQmCC","orcid":"","institution":"Chinese National Information Technology Research Institute, Northwest Minzu University","correspondingAuthor":true,"prefix":"","firstName":"Wenjin","middleName":"","lastName":"Hu","suffix":""},{"id":490304701,"identity":"1c01a5bd-5494-413d-a8ed-102ef5551e60","order_by":2,"name":"Yinqiu Zhao","email":"","orcid":"","institution":"Chinese National Information Technology Research Institute, Northwest Minzu University","correspondingAuthor":false,"prefix":"","firstName":"Yinqiu","middleName":"","lastName":"Zhao","suffix":""},{"id":490304704,"identity":"7fb8868b-2c73-42f7-bd06-10d15b21a545","order_by":3,"name":"Fujun Zhang","email":"","orcid":"","institution":"Chinese National Information Technology Research Institute, Northwest Minzu University","correspondingAuthor":false,"prefix":"","firstName":"Fujun","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2025-07-12 12:08:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7108021/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7108021/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s40494-026-02343-8","type":"published","date":"2026-02-07T15:57:18+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":102235591,"identity":"d3b94ce8-75fd-4157-8a72-a9e47d66be8d","added_by":"auto","created_at":"2026-02-09 16:17:05","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1828117,"visible":true,"origin":"","legend":"","description":"","filename":"GeoTCAMAThangkaCaptioningMethodIntegratingTopicModelingwithGeometryGuidedSpatialAttention.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7108021/v1_covered_a09f17d4-7aed-4783-a576-3f32e50efa16.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Geo-TCAM: A Thangka Captioning Method Integrating Topic Modeling with Geometry- Guided Spatial Attention","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":"npj-heritage-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hsci","sideBox":"Learn more about [Heritage Science](http://heritagesciencejournal.springeropen.com)","snPcode":"40494","submissionUrl":"https://submission.nature.com/new-submission/40494/3","title":"npj Heritage Science","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7108021/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7108021/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Thangka image captioning, an essential task in cultural heritage preservation, faces challenges due to the complexity of Thangka imagery and the depth of their semantic content. Current deep learning-based methods struggle with extracting detailed features and accurately understanding the semantics of Thangka images, often leading to incomplete or incorrect captions of key elements such as the main deity and the background. To address these challenges, this paper introduces a novel Thangka captioning model, integrating topic modeling and geometry-guided spatial attention (Geo-TCAM). The model employs a multi-level feature integration strategy to enhance feature extraction, including gestures and objects. By incorporating Latent Dirichlet Allocation (LDA) topic weights and visual features (TIF), it leverages external domain knowledge for better semantic understanding. The Geo-TCAM's geometry-guided facial spatial attention module (GFSA) improves spatial layout recognition. Experimental results demonstrate significant improvements in captioning performance, with BLEU-1, BLEU-4, METEOR, and CIDEr scores increasing by 11.9%, 17.1%, 9.7%, and 119.5%, respectively, compared to baseline models. On the COCO public dataset, the Geo-TCAM model also demonstrates outstanding performance, comparable to that of other state-of-the-art models. This study employs the Geo-TCAM model to accurately generate image captions for Thangka images, facilitating the digital preservation and dissemination of cultural heritage.","manuscriptTitle":"Geo-TCAM: A Thangka Captioning Method Integrating Topic Modeling with Geometry- Guided Spatial Attention","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-25 03:00:54","doi":"10.21203/rs.3.rs-7108021/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-13T00:50:58+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-12T06:30:37+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-11T14:24:30+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-07T07:53:39+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"339850005130030779902545995822547819515","date":"2025-07-29T00:41:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"137274535373268767874873569194940282135","date":"2025-07-27T22:38:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"331214548237933235466256953631119007925","date":"2025-07-25T16:27:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"113437058957983168379440888360563360853","date":"2025-07-24T11:43:20+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-24T07:31:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"317703089965579401681618714183623576262","date":"2025-07-23T08:58:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"200009065828066883509103181431198859804","date":"2025-07-23T08:57:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"182301029186580539625684807950958448657","date":"2025-07-23T08:57:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"46809643027351567709803710465801533378","date":"2025-07-23T08:54:29+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-23T08:48:32+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-17T06:40:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-17T06:37:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj heritage science","date":"2025-07-12T11:56:34+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-heritage-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hsci","sideBox":"Learn more about [Heritage Science](http://heritagesciencejournal.springeropen.com)","snPcode":"40494","submissionUrl":"https://submission.nature.com/new-submission/40494/3","title":"npj Heritage Science","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"72454ec9-8360-4410-bd7b-bdbe2d114371","owner":[],"postedDate":"July 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-02-09T16:16:06+00:00","versionOfRecord":{"articleIdentity":"rs-7108021","link":"https://doi.org/10.1038/s40494-026-02343-8","journal":{"identity":"npj-heritage-science","isVorOnly":false,"title":"npj Heritage Science"},"publishedOn":"2026-02-07 15:57:18","publishedOnDateReadable":"February 7th, 2026"},"versionCreatedAt":"2025-07-25 03:00:54","video":"","vorDoi":"10.1038/s40494-026-02343-8","vorDoiUrl":"https://doi.org/10.1038/s40494-026-02343-8","workflowStages":[]},"version":"v1","identity":"rs-7108021","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7108021","identity":"rs-7108021","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}