Epipolar Constraint Guided Differentiable Keypoint Detection and Description

Epipolar Constraint Guided Differentiable Keypoint Detection and Description | 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 Epipolar Constraint Guided Differentiable Keypoint Detection and Description Xi Li, Yulong Feng, Xianguo Yu, Yirui Cong, Lili Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4410154/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Feb, 2025 Read the published version in The Visual Computer → Version 1 posted 13 You are reading this latest preprint version Abstract Sparse local feature matching methods have made significant strides in a variety of visual geometric tasks. Recently, the weakly supervised under epipolar constraint approaches achieve stronger performance compared to the fully supervised methods through the decoupled training describe-then-detect. However, such methods base on the policy gradient to train the detector, ignoring the reliability of the keypoints. Meanwhile, many of the sparse local feature matching methods place more emphasis on accuracy over speed, being unfriendly for real-time applications. To address these issues, we introduce the differentiable keypoint extraction and the dispersity peak loss to generate clean score maps and enhance the reliability of the keypoints in the weakly supervised method. Additionally, we propose a straightforward yet efficient model that achieves a good balance between accuracy and speed. We conduct experiments on multiple public benchmark datasets, achieving higher performance than existing methods. The code will be available at https://github.com/FYL0123/WSDK . Local feature matching Weakly supervised Epipolar constraint Differentiable keypoint extraction Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 09 Feb, 2025 Read the published version in The Visual Computer → Version 1 posted Editorial decision: Revision requested 09 Dec, 2024 Reviewers agreed at journal 08 Dec, 2024 Reviews received at journal 08 Dec, 2024 Reviews received at journal 08 Dec, 2024 Reviews received at journal 06 Dec, 2024 Reviewers agreed at journal 06 Dec, 2024 Reviewers agreed at journal 05 Dec, 2024 Reviewers agreed at journal 05 Dec, 2024 Reviewers agreed at journal 28 May, 2024 Reviewers invited by journal 22 May, 2024 Editor assigned by journal 13 May, 2024 Submission checks completed at journal 13 May, 2024 First submitted to journal 12 May, 2024 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-4410154","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":304178459,"identity":"1d9199b8-8aba-4135-8030-5ed8f251f2de","order_by":0,"name":"Xi Li","email":"","orcid":"","institution":"Changsha University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Xi","middleName":"","lastName":"Li","suffix":""},{"id":304178460,"identity":"be17f747-0240-44b9-a482-3e22977c9aa2","order_by":1,"name":"Yulong Feng","email":"","orcid":"","institution":"Changsha University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Yulong","middleName":"","lastName":"Feng","suffix":""},{"id":304178461,"identity":"2f3a7755-bf0d-4125-af71-10a308a73bb8","order_by":2,"name":"Xianguo Yu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYJCCA2CSmbHxAZAEMQ2I1MLO3AxUyixBlBYI4GdvkyBKi8GNHMMDP3fUJvYzM7ZVF7ZZ1zGwN2+TYKi5g0dLWsLB3jPHE2c2M7bdntmWLsHAc6xMguHYMzxakg8c4G07lrjhMFALb9thCQaJHDMJxobDeLQkNhz8C9VSDNYi/4aQluQDh3nbasBamCG28ODXInnmWcJh2bYDxkC/NEvznEuXbONJK7ZIOIZbC9/xHOOPb9vqZPv5jz/8zFNmzc/PfnjjjQ81uLUoHABTSArYQEQCTg0MDPINYKoOj5JRMApGwSgY8QAARWBX3Xa6iVoAAAAASUVORK5CYII=","orcid":"","institution":"National University of Defense Technology","correspondingAuthor":true,"prefix":"","firstName":"Xianguo","middleName":"","lastName":"Yu","suffix":""},{"id":304178463,"identity":"d97dd886-8085-4066-bfa1-0be63fa862c6","order_by":3,"name":"Yirui Cong","email":"","orcid":"","institution":"National University of Defense Technology","correspondingAuthor":false,"prefix":"","firstName":"Yirui","middleName":"","lastName":"Cong","suffix":""},{"id":304178464,"identity":"d8fd5cfc-f020-420c-8ea4-06d89395c742","order_by":4,"name":"Lili Chen","email":"","orcid":"","institution":"Hunan Industry Polytechnic","correspondingAuthor":false,"prefix":"","firstName":"Lili","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2024-05-13 02:02:37","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4410154/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4410154/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00371-024-03795-4","type":"published","date":"2025-02-09T15:57:17+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":75931333,"identity":"f4ea0dd4-6a7e-45b4-9f25-de70445fb2e1","added_by":"auto","created_at":"2025-02-10 16:14:35","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4619846,"visible":true,"origin":"","legend":"","description":"","filename":"snarticletemplate.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4410154/v1_covered_0f81abae-34ed-4be9-acfe-026a4efe197c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Epipolar Constraint Guided Differentiable Keypoint Detection and Description","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":"the-visual-computer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"tvcj","sideBox":"Learn more about [The Visual Computer](http://link.springer.com/journal/371)","snPcode":"371","submissionUrl":"https://submission.nature.com/new-submission/371/3","title":"The Visual Computer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Local feature matching, Weakly supervised, Epipolar constraint, Differentiable keypoint extraction","lastPublishedDoi":"10.21203/rs.3.rs-4410154/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4410154/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":" Sparse local feature matching methods have made significant strides in a variety of visual geometric tasks. Recently, the weakly supervised under epipolar constraint approaches achieve stronger performance compared to the fully supervised methods through the decoupled training describe-then-detect. However, such methods base on the policy gradient to train the detector, ignoring the reliability of the keypoints. Meanwhile, many of the sparse local feature matching methods place more emphasis on accuracy over speed, being unfriendly for real-time applications. To address these issues, we introduce the differentiable keypoint extraction and the dispersity peak loss to generate clean score maps and enhance the reliability of the keypoints in the weakly supervised method. Additionally, we propose a straightforward yet efficient model that achieves a good balance between accuracy and speed. We conduct experiments on multiple public benchmark datasets, achieving higher performance than existing methods. The code will be available at https://github.com/FYL0123/WSDK.","manuscriptTitle":"Epipolar Constraint Guided Differentiable Keypoint Detection and Description","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-05-22 02:12:57","doi":"10.21203/rs.3.rs-4410154/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-12-09T21:16:10+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"261186782599501805080920531173836634363","date":"2024-12-09T04:06:20+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-09T03:03:48+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-08T22:13:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-06T07:36:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"135789916660594534572587158644268200449","date":"2024-12-06T06:01:06+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"56201160552378066009976264710525566237","date":"2024-12-06T03:38:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"148710252847173725192091373323029295110","date":"2024-12-06T03:05:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"109420688227178034186137403908407710440","date":"2024-05-29T01:46:33+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-05-22T12:23:54+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-05-13T11:20:12+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-05-13T06:33:10+00:00","index":"","fulltext":""},{"type":"submitted","content":"The Visual Computer","date":"2024-05-13T02:01:16+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-visual-computer","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"tvcj","sideBox":"Learn more about [The Visual Computer](http://link.springer.com/journal/371)","snPcode":"371","submissionUrl":"https://submission.nature.com/new-submission/371/3","title":"The Visual Computer","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"f7508884-ce1d-40a8-8bb9-917d06e0fc3a","owner":[],"postedDate":"May 22nd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-02-10T16:10:24+00:00","versionOfRecord":{"articleIdentity":"rs-4410154","link":"https://doi.org/10.1007/s00371-024-03795-4","journal":{"identity":"the-visual-computer","isVorOnly":false,"title":"The Visual Computer"},"publishedOn":"2025-02-09 15:57:17","publishedOnDateReadable":"February 9th, 2025"},"versionCreatedAt":"2024-05-22 02:12:57","video":"","vorDoi":"10.1007/s00371-024-03795-4","vorDoiUrl":"https://doi.org/10.1007/s00371-024-03795-4","workflowStages":[]},"version":"v1","identity":"rs-4410154","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4410154","identity":"rs-4410154","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}