Open-Vocabulary 3D Understanding with Identity-Enhanced Segmentation

Open-Vocabulary 3D Understanding with Identity-Enhanced Segmentation | 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 Open-Vocabulary 3D Understanding with Identity-Enhanced Segmentation Weijie Lin, Wei Xiang, Lu Yu, Tianyu Chen, Kang Han This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6424450/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 Open-vocabulary 3D segmentation assigns semantic labels to 3D data by matching language embeddings with 3D rendered semantic vision-language embeddings. To achieve this function, existing methods fuse 2D semantic embeddings extracted from multi-view images by 2D foundation models into a unified 3D representation. However, these methods often struggle when multi-view images have significant viewpoint changes, object deformations, and occlusions, leading to inconsistent object identities and reduced segmentation accuracy. To overcome these challenges, we propose a novel framework, 3D Identity-Enhanced Segmentation (3D-IES). 3D-IES leverages multi-view geometry and a fully trained 3D Gaussian Splatting model to reproject 2D segmentation masks into the 3D space, thereby enabling consistent object identity assignment across diverse viewpoints. By anchoring segmentation masks in the 3D space, our method ensures spatial consistency, robust object tracking, and accurate segmentation, even under challenging conditions such as significant viewpoint changes or overlapping regions. Experimental results demonstrate that 3D-IES significantly outperforms state-of-the-art methods in open-vocabulary 3D semantic segmentation, achieving superior robustness and accuracy across a variety of complex scenes. Physical sciences/Mathematics and computing/Computational science Physical sciences/Mathematics and computing/Computer science Physical sciences/Mathematics and computing/Information technology 3D segmentation Open-vocabulary 3D Gaussian splatting Semantic scene understanding 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-6424450","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":452892425,"identity":"d7448ded-aa98-49ba-b02d-2feb19351af9","order_by":0,"name":"Weijie Lin","email":"","orcid":"","institution":"La Trobe University","correspondingAuthor":false,"prefix":"","firstName":"Weijie","middleName":"","lastName":"Lin","suffix":""},{"id":452892432,"identity":"2b9a43cc-827a-4d8e-aee2-3def1696ec79","order_by":1,"name":"Wei Xiang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYJCCA0AsxwdiMTaAyATitBizMTCToAUEEtuI1mJwgMfwwM8dtelt7OePffi4w4aBnz3HgOFnGz4tbAkHe88cz23jSWaeOfNMGoNkzxsDxl48WswOMB84wNt2LLeNIZmZmbftMIPBDaAtvHi1MDYc/Nt2LJ2N/zFIy38Ge6AWxr8EbDnM21aTwCYBtuUAg4FEjgEzPlvsD7MlHJZtO2DYJvHYmHHmmWQeiTPPCg7LnMOtRbK9x/jj27Y6eX7+xMcMH3fYyfG3J298+KYMtxZQbADBYTifB0QcwKMBBuqIUDMKRsEoGAUjFgAAFQxPkVN420IAAAAASUVORK5CYII=","orcid":"","institution":"La Trobe University","correspondingAuthor":true,"prefix":"","firstName":"Wei","middleName":"","lastName":"Xiang","suffix":""},{"id":452892433,"identity":"9458c11d-109e-444d-9bd2-6cbdef7511ed","order_by":2,"name":"Lu Yu","email":"","orcid":"","institution":"La Trobe University","correspondingAuthor":false,"prefix":"","firstName":"Lu","middleName":"","lastName":"Yu","suffix":""},{"id":452892434,"identity":"a092f017-1a2f-42e3-a786-5857f1e0bb0b","order_by":3,"name":"Tianyu Chen","email":"","orcid":"","institution":"La Trobe University","correspondingAuthor":false,"prefix":"","firstName":"Tianyu","middleName":"","lastName":"Chen","suffix":""},{"id":452892435,"identity":"4ee3b9ea-d78b-4d12-a706-e52f900fdfe5","order_by":4,"name":"Kang Han","email":"","orcid":"","institution":"La Trobe University","correspondingAuthor":false,"prefix":"","firstName":"Kang","middleName":"","lastName":"Han","suffix":""}],"badges":[],"createdAt":"2025-04-11 03:53:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6424450/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6424450/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90289740,"identity":"db9552a8-4dda-4025-80be-f328591caf3f","added_by":"auto","created_at":"2025-09-01 07:09:35","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6965107,"visible":true,"origin":"","legend":"","description":"","filename":"OpenVocabulary3DUnderstandingwithIdentityEnhancedSegmentation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6424450/v1_covered_1d7b2392-1b7e-44cb-82a0-fc8752fdd0a4.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Open-Vocabulary 3D Understanding with Identity-Enhanced Segmentation","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":"3D segmentation, Open-vocabulary, 3D Gaussian splatting, Semantic scene understanding","lastPublishedDoi":"10.21203/rs.3.rs-6424450/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6424450/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Open-vocabulary 3D segmentation assigns semantic labels to 3D data by matching language embeddings with 3D rendered semantic vision-language embeddings. To achieve this function, existing methods fuse 2D semantic embeddings extracted from multi-view images by 2D foundation models into a unified 3D representation. However, these methods often struggle when multi-view images have significant viewpoint changes, object deformations, and occlusions, leading to inconsistent object identities and reduced segmentation accuracy. To overcome these challenges, we propose a novel framework, 3D Identity-Enhanced Segmentation (3D-IES). 3D-IES leverages multi-view geometry and a fully trained 3D Gaussian Splatting model to reproject 2D segmentation masks into the 3D space, thereby enabling consistent object identity assignment across diverse viewpoints. By anchoring segmentation masks in the 3D space, our method ensures spatial consistency, robust object tracking, and accurate segmentation, even under challenging conditions such as significant viewpoint changes or overlapping regions. Experimental results demonstrate that 3D-IES significantly outperforms state-of-the-art methods in open-vocabulary 3D semantic segmentation, achieving superior robustness and accuracy across a variety of complex scenes.","manuscriptTitle":"Open-Vocabulary 3D Understanding with Identity-Enhanced Segmentation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-07 16:12:15","doi":"10.21203/rs.3.rs-6424450/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":"81f0b7dd-9d3d-4f15-a5d6-bc914b853e8c","owner":[],"postedDate":"May 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48165202,"name":"Physical sciences/Mathematics and computing/Computational science"},{"id":48165203,"name":"Physical sciences/Mathematics and computing/Computer science"},{"id":48165204,"name":"Physical sciences/Mathematics and computing/Information technology"}],"tags":[],"updatedAt":"2025-09-01T07:08:18+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-07 16:12:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6424450","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6424450","identity":"rs-6424450","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}