OmniGlasses: An Optical Aid for Stereo Vision CNNs to Enable Omnidirectional Image Processing

OmniGlasses: An Optical Aid for Stereo Vision CNNs to Enable Omnidirectional Image Processing | 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 OmniGlasses: An Optical Aid for Stereo Vision CNNs to Enable Omnidirectional Image Processing Julian Bruno Seuffert, Ana Cecilia Perez Grassi, Hamza Ahmed, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2776786/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 Apr, 2024 Read the published version in Machine Vision and Applications → Version 1 posted You are reading this latest preprint version Abstract Stereo vision is a key technology for 3D scene reconstruction from image pairs. Most approaches process perspective images from commodity cameras. These images, however, have a very limited field of view and only picture a small portion of the scene. In contrast, omnidirectional images, also known as fisheye images, exhibit a much larger field of view and allow a full 3D scene reconstruction with a small amount of cameras if placed carefully. However, omnidirectional images are strongly distorted which make the 3D reconstruction much more sophisticated. Nowadays, a lot of research is conducted on CNNs for omnidirectional stereo vision. Nevertheless, a significant gap between estimation accuracy and throughput can be observed in the literature. This work aims to bridge this gap by introducing a novel set of transformations, namely OmniGlasses . These are incorporated into the architecture of a fast network, i.e., AnyNet , originally designed for scene reconstruction on perspective images. Our network, Omni-AnyNet , produces accurate omnidirectional distance maps with a mean absolute error of around 13 cm at 36.4 fps and is therefore real-time capable. Artificial Intelligence and Machine Learning Epipolar Geometry Fisheye Omnidirectional Look Up Table Stereo Vision View Synthesis Full Text Cite Share Download PDF Status: Published Journal Publication published 23 Apr, 2024 Read the published version in Machine Vision and Applications → 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-2776786","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":189105697,"identity":"89f03106-0301-423a-839a-c0e24170db5b","order_by":0,"name":"Julian Bruno Seuffert","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-0636-3385","institution":"Chemnitz University of Technology","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Julian","middleName":"Bruno","lastName":"Seuffert","suffix":""},{"id":189105698,"identity":"c77ba2f2-e47c-4b68-8b17-36c46b1caa7a","order_by":1,"name":"Ana Cecilia Perez Grassi","email":"","orcid":"https://orcid.org/0000-0003-1171-903X","institution":"Chemnitz University of Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ana","middleName":"Cecilia Perez","lastName":"Grassi","suffix":""},{"id":189105699,"identity":"e105c6b3-c990-49dd-a26d-6a5183f06d2b","order_by":2,"name":"Hamza Ahmed","email":"","orcid":"https://orcid.org/0000-0002-6038-3328","institution":"Chemnitz University of Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hamza","middleName":"","lastName":"Ahmed","suffix":""},{"id":189105700,"identity":"57663c18-a0b1-4016-83bf-2906f3a04e3b","order_by":3,"name":"Roman Seidel","email":"","orcid":"https://orcid.org/0000-0002-3144-1488","institution":"Chemnitz University of Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Roman","middleName":"","lastName":"Seidel","suffix":""},{"id":189105701,"identity":"de84d4df-c7c7-4a10-98e8-a73423d6f5b9","order_by":4,"name":"Gangolf Hirtz","email":"","orcid":"https://orcid.org/0000-0002-4393-5354","institution":"Chemnitz University of Technology","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Gangolf","middleName":"","lastName":"Hirtz","suffix":""}],"badges":[],"createdAt":"2023-04-04 13:28:59","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":true,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false,"coiExplicitlySet":false},"doi":"10.21203/rs.3.rs-2776786/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2776786/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00138-024-01534-2","type":"published","date":"2024-04-23T22:43:21+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":55690643,"identity":"c33431e6-62fd-457f-a465-517b2dc71900","added_by":"auto","created_at":"2024-05-01 22:43:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2046995,"visible":true,"origin":"","legend":"","description":"","filename":"snarticle.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2776786/v1_covered_707bb092-958e-45be-8bbf-736b4ab82e5f.pdf"}],"financialInterests":"","formattedTitle":"\u003cp\u003eOmniGlasses: An Optical Aid for Stereo Vision CNNs to Enable Omnidirectional Image Processing\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"Chemnitz University of Technology","isAcceptedByJournal":true,"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":"Epipolar Geometry, Fisheye, Omnidirectional, Look Up Table, Stereo Vision, View Synthesis","lastPublishedDoi":"10.21203/rs.3.rs-2776786/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2776786/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eStereo vision is a key technology for 3D scene reconstruction from image pairs. Most approaches process perspective images from commodity cameras. These images, however, have a very limited field of view and only picture a small portion of the scene. In contrast, omnidirectional images, also known as fisheye images, exhibit a much larger field of view and allow a full 3D scene reconstruction with a small amount of cameras if placed carefully. However, omnidirectional images are strongly distorted which make the 3D reconstruction much more sophisticated. Nowadays, a lot of research is conducted on CNNs for omnidirectional stereo vision. Nevertheless, a significant gap between estimation accuracy and throughput can be observed in the literature. This work aims to bridge this gap by introducing a novel set of transformations, namely \u003cem\u003eOmniGlasses\u003c/em\u003e. These are incorporated into the architecture of a fast network, i.e., \u003cem\u003eAnyNet\u003c/em\u003e, originally designed for scene reconstruction on perspective images. Our network, \u003cem\u003eOmni-AnyNet\u003c/em\u003e, produces accurate omnidirectional distance maps with a mean absolute error of around \u003cstrong\u003e13\u003c/strong\u003e cm at \u003cstrong\u003e36.4\u003c/strong\u003e fps and is therefore real-time capable.\u003c/p\u003e","manuscriptTitle":"OmniGlasses: An Optical Aid for Stereo Vision CNNs to Enable Omnidirectional Image Processing","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2023-04-05 03:37:45","doi":"10.21203/rs.3.rs-2776786/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":"8e89f17e-9056-40eb-9a34-77ad36cdf5f8","owner":[],"postedDate":"April 5th, 2023","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":20451271,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2024-05-01T22:43:21+00:00","versionOfRecord":{"articleIdentity":"rs-2776786","link":"https://doi.org/10.1007/s00138-024-01534-2","journal":{"identity":"machine-vision-and-applications","isVorOnly":false,"title":"Machine Vision and Applications"},"publishedOn":"2024-04-23 22:43:21","publishedOnDateReadable":"April 23rd, 2024"},"versionCreatedAt":"2023-04-05 03:37:45","video":"","vorDoi":"10.1007/s00138-024-01534-2","vorDoiUrl":"https://doi.org/10.1007/s00138-024-01534-2","workflowStages":[]},"version":"v1","identity":"rs-2776786","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2776786","identity":"rs-2776786","version":["v1"]},"buildId":"WrCJVZZCHTDjtuVLN7oU0","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}