Deep Stereo Network with Cross-Correlation Volume Construction and Least Square Aggregation

Deep Stereo Network with Cross-Correlation Volume Construction and Least Square Aggregation | 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 Deep Stereo Network with Cross-Correlation Volume Construction and Least Square Aggregation Guowei An, Yaonan Wang, Qing Zhu, Xiaofang Yuan, Kai Zeng, Yang Mo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3857366/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 Stereo matching is of great importance in robot operation, autonomous driving and virtual reality. Large textureless and depth discontinuity regions are still the problems that limit the accuracy of stereo matching tasks. Traditional correlation-based volumes only measure the feature similarity within the same channel of the feature maps, which lead to insufficient feature similarity learning between different channels and perform poorly in large textureless regions with high feature similarity requirements. To address the problems in large textureless regions, we propose the cross-correlation based cost volume construction which adequately learn the feature similarity in different channels of the feature maps. To address the problems in depth discontinuity regions and other gradient sensitive regions, we propose the differentiable least square aggregation module which can sufficiently utilize the gradient information and enhance the aggregation ability of the cost aggregation network for gradient features. Extensive experiments show that the proposed method solve the problems effectively in the above difficult regions and achieve the state-of-the art performance on Scene Flow dataset, KITTI 2012 dataset and KITTI 2015 dataset. stereo matching volume construction neural network aggregation stereo vision. Full Text 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-3857366","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":288534808,"identity":"2b7bb68d-4831-4f58-9357-f7682ff42245","order_by":0,"name":"Guowei An","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqElEQVRIiWNgGAWjYHACxscQOoF4LczGJGthkyZNC3//GrPqwh2HGfjZcwwYfu4gQovEjTdmt2eeOcwg2fPGgLH3DDHW3Di77TZv22EGgxs5BsyMbUTokAdqKQZpsSdai8H53m3MYFskiNVieIP/szRvWzqPxJlnBQd7idEid/5Y4mfeNms5/vbkjQ9+EqOFQSIBTPGAiAPEaABGDJHqRsEoGAWjYAQDAB0zNGLyKqT4AAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-6490-4277","institution":"Hunan University","correspondingAuthor":true,"prefix":"","firstName":"Guowei","middleName":"","lastName":"An","suffix":""},{"id":288534809,"identity":"82751dd9-34a4-48d0-9ad0-1b92301892fc","order_by":1,"name":"Yaonan Wang","email":"","orcid":"","institution":"Hunan University","correspondingAuthor":false,"prefix":"","firstName":"Yaonan","middleName":"","lastName":"Wang","suffix":""},{"id":288534810,"identity":"15ebe270-ee70-46e7-a0ec-54d508486100","order_by":2,"name":"Qing Zhu","email":"","orcid":"","institution":"Hunan University","correspondingAuthor":false,"prefix":"","firstName":"Qing","middleName":"","lastName":"Zhu","suffix":""},{"id":288534811,"identity":"fe432805-ea20-46ad-8ed5-cd4832cf683a","order_by":3,"name":"Xiaofang Yuan","email":"","orcid":"","institution":"Hunan University","correspondingAuthor":false,"prefix":"","firstName":"Xiaofang","middleName":"","lastName":"Yuan","suffix":""},{"id":288534812,"identity":"8567b1bc-962a-43f2-9c4b-1b1a40262190","order_by":4,"name":"Kai Zeng","email":"","orcid":"","institution":"Hunan University","correspondingAuthor":false,"prefix":"","firstName":"Kai","middleName":"","lastName":"Zeng","suffix":""},{"id":288534813,"identity":"4e410aeb-ad8d-48f5-9d0e-90fbb5fbfd0f","order_by":5,"name":"Yang Mo","email":"","orcid":"","institution":"Hunan University","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Mo","suffix":""}],"badges":[],"createdAt":"2024-01-12 15:10:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3857366/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3857366/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":63834625,"identity":"2e8877d8-5409-42ee-a588-b9fcabb872e3","added_by":"auto","created_at":"2024-09-02 20:16:16","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1393584,"visible":true,"origin":"","legend":"","description":"","filename":"manuscriptoftheDCLS.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3857366/v1_covered_bfeb022c-6eca-4b48-b60a-fc73aae52ee3.pdf"}],"financialInterests":"","formattedTitle":"Deep Stereo Network with Cross-Correlation Volume Construction and Least Square Aggregation","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":"stereo matching, volume construction, neural network, aggregation, stereo vision.","lastPublishedDoi":"10.21203/rs.3.rs-3857366/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3857366/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Stereo matching is of great importance in robot operation, autonomous driving and virtual reality. Large textureless and depth discontinuity regions are still the problems that limit the accuracy of stereo matching tasks. Traditional correlation-based volumes only measure the feature similarity within the same channel of the feature maps, which lead to insufficient feature similarity learning between different channels and perform poorly in large textureless regions with high feature similarity requirements. To address the problems in large textureless regions, we propose the cross-correlation based cost volume construction which adequately learn the feature similarity in different channels of the feature maps. To address the problems in depth discontinuity regions and other gradient sensitive regions, we propose the differentiable least square aggregation module which can sufficiently utilize the gradient information and enhance the aggregation ability of the cost aggregation network for gradient features. Extensive experiments show that the proposed method solve the problems effectively in the above difficult regions and achieve the state-of-the art performance on Scene Flow dataset, KITTI 2012 dataset and KITTI 2015 dataset.","manuscriptTitle":"Deep Stereo Network with Cross-Correlation Volume Construction and Least Square Aggregation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-10 06:16:44","doi":"10.21203/rs.3.rs-3857366/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":"53d6d626-f558-4d8b-8952-3e1f7d1d69a4","owner":[],"postedDate":"April 10th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-09-02T20:08:09+00:00","versionOfRecord":[],"versionCreatedAt":"2024-04-10 06:16:44","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3857366","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3857366","identity":"rs-3857366","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}