Target-Aware Local Mesh Repair for Semantic Part Removal with Minimal Geometric Change

Target-Aware Local Mesh Repair for Semantic Part Removal with Minimal Geometric Change | 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 Target-Aware Local Mesh Repair for Semantic Part Removal with Minimal Geometric Change Bingyang Ji, Changxin Gao, Fuhao Chen, Shan Cui This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9492440/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Mesh repair and hole filling are fundamental tasks in geometric processing and computer graphics. Traditional hole-filling methods address generic holes, whereas 3D shape completion aims to reconstruct full objects. We introduce a new task: minimal-change local mesh repair after semantic part removal, which closes only the opening caused by part deletion while preserving unaffected regions. We build a reproducible PartNet-derived benchmark with a primary chair-leg split, supplementary cross-category extensions on tables and storage furniture. To identify the genuine repair target, we propose a removed-part-aware boundary loop selection scheme, and further explore a lightweight learning-based loop classifier (Random Forest / MLP / GBDT) based on geometric features. We evaluate two geometric baselines (center-fan and planar triangulation), an advancing-front method, a generic fill-all baseline, and a Poisson reconstruction baseline using closure, patch complexity, triangle quality, locality, and geometric distance. Experiments show that accurate target selection is more important than the specific triangulation backend, that planar triangulation produces cleaner patches without extra vertices, and that the target-aware formulation generalizes beyond the chair benchmark and remains beneficial in preliminary multi-part settings. The source code, dataset construction pipeline, versioned release, and evaluation benchmark are publicly available at \url{ https://github.com/jibingyang11/3D-Part-Repair} ; the archived release is available at \url{ https://doi.org/10.5281/zenodo.19533166} . Physical sciences/Engineering Physical sciences/Mathematics and computing Mesh repair Semantic part removal Local repair Target-aware selection Geometric processing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 22 May, 2026 Reviews received at journal 19 May, 2026 Reviewers agreed at journal 07 May, 2026 Reviewers invited by journal 07 May, 2026 Editor invited by journal 24 Apr, 2026 Editor assigned by journal 22 Apr, 2026 Submission checks completed at journal 22 Apr, 2026 First submitted to journal 22 Apr, 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-9492440","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":641543856,"identity":"b9428010-6182-40ac-acf3-18be149edff6","order_by":0,"name":"Bingyang Ji","email":"","orcid":"","institution":"Linyi University","correspondingAuthor":false,"prefix":"","firstName":"Bingyang","middleName":"","lastName":"Ji","suffix":""},{"id":641543857,"identity":"a363fa34-d9e6-4187-b832-b069b44c3d3b","order_by":1,"name":"Changxin Gao","email":"","orcid":"","institution":"Linyi University","correspondingAuthor":false,"prefix":"","firstName":"Changxin","middleName":"","lastName":"Gao","suffix":""},{"id":641543858,"identity":"0b23cb93-697d-4895-af7d-16ae5dd24c01","order_by":2,"name":"Fuhao Chen","email":"","orcid":"","institution":"Linyi University","correspondingAuthor":false,"prefix":"","firstName":"Fuhao","middleName":"","lastName":"Chen","suffix":""},{"id":641543859,"identity":"59fde699-93f3-4b57-b748-ae2d58c55ccd","order_by":3,"name":"Shan Cui","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyElEQVRIiWNgGAWjYFACxgZmICnHB+awkaDFmI0ELQwMIC2JbURr4Z+R3Py5oOZOeptEjgHDh7LDDPyzG/BrkbiR2CY949izXJAWxhnnDjNI3DmAX4uBRGIbMw/bYbAWZt62w0CRBIJamj/z/DuczgbS8pdILQ3SQMMTwFoYidEiceZhmzRv32HDNp5nBQd7zqXzSNwgoIW/Pf3xZ55vh+X52ZM3PvhRZi3HP4OAFgQQSGA4AKR4iFUPsu8ACYpHwSgYBaNgRAEA0xc+DFKE+H4AAAAASUVORK5CYII=","orcid":"","institution":"Linyi University","correspondingAuthor":true,"prefix":"","firstName":"Shan","middleName":"","lastName":"Cui","suffix":""}],"badges":[],"createdAt":"2026-04-22 07:26:52","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9492440/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9492440/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109759486,"identity":"1a929026-c1ee-4efc-b1f5-81c0790fb3c3","added_by":"auto","created_at":"2026-05-22 07:27:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5951178,"visible":true,"origin":"","legend":"","description":"","filename":"paper.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9492440/v1_covered_07b9e7a7-0667-45ba-a4e1-0dbdd57edd6f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Target-Aware Local Mesh Repair for Semantic Part Removal with Minimal Geometric Change","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"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":"Mesh repair, Semantic part removal, Local repair, Target-aware selection, Geometric processing","lastPublishedDoi":"10.21203/rs.3.rs-9492440/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9492440/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Mesh repair and hole filling are fundamental tasks in geometric processing and computer graphics. Traditional hole-filling methods address generic holes, whereas 3D shape completion aims to reconstruct full objects. We introduce a new task: minimal-change local mesh repair after semantic part removal, which closes only the opening caused by part deletion while preserving unaffected regions. We build a reproducible PartNet-derived benchmark with a primary chair-leg split, supplementary cross-category extensions on tables and storage furniture. To identify the genuine repair target, we propose a removed-part-aware boundary loop selection scheme, and further explore a lightweight learning-based loop classifier (Random Forest / MLP / GBDT) based on geometric features. We evaluate two geometric baselines (center-fan and planar triangulation), an advancing-front method, a generic fill-all baseline, and a Poisson reconstruction baseline using closure, patch complexity, triangle quality, locality, and geometric distance. Experiments show that accurate target selection is more important than the specific triangulation backend, that planar triangulation produces cleaner patches without extra vertices, and that the target-aware formulation generalizes beyond the chair benchmark and remains beneficial in preliminary multi-part settings. The source code, dataset construction pipeline, versioned release, and evaluation benchmark are publicly available at \\url{https://github.com/jibingyang11/3D-Part-Repair}; the archived release is available at \\url{https://doi.org/10.5281/zenodo.19533166}.","manuscriptTitle":"Target-Aware Local Mesh Repair for Semantic Part Removal with Minimal Geometric Change","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-18 15:14:55","doi":"10.21203/rs.3.rs-9492440/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"270084927652832874855800137721023779261","date":"2026-05-22T06:44:55+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-19T21:49:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"47002454017368678443015835298372346560","date":"2026-05-07T14:49:08+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-05-07T13:52:46+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-04-24T10:11:16+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-22T10:26:29+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-22T10:26:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-04-22T07:21:16+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":"eeeed086-c017-43e7-a6f4-b10351f19c36","owner":[],"postedDate":"May 18th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"270084927652832874855800137721023779261","date":"2026-05-22T06:44:55+00:00","index":38,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-19T21:49:58+00:00","index":37,"fulltext":""},{"type":"reviewerAgreed","content":"47002454017368678443015835298372346560","date":"2026-05-07T14:49:08+00:00","index":32,"fulltext":""},{"type":"reviewersInvited","content":"7","date":"2026-05-07T13:52:46+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":68240036,"name":"Physical sciences/Engineering"},{"id":68240037,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-18T15:14:55+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-18 15:14:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9492440","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9492440","identity":"rs-9492440","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}