A Deep Learning-Based Visual SLAM Approach for Indoor Dynamic Scenes

A Deep Learning-Based Visual SLAM Approach for Indoor Dynamic Scenes | 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 A Deep Learning-Based Visual SLAM Approach for Indoor Dynamic Scenes jiarui qin, yugang wang, xueli cong, liyao zhou This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8981127/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 21 You are reading this latest preprint version Abstract In indoor dynamic scenes, traditional visual SLAM algorithms often suffer from a significant decline in localization accuracy due to interference from dynamic objects. To address this issue, this study proposes a visual SLAM method based on deep learning. Built upon the ORB-SLAM3 framework, the proposed method effectively enhances the system's capability in complex dynamic environments by constructing a dual dynamic feature suppression mechanism that integrates object detection and epipolar geometric constraints. Specifically, the YOLOv8m network is first lightweighted by introducing GhostConv modules to replace the original convolutional layers, reducing computational overhead. Additionally, the CBAM attention mechanism is incorporated into the backbone network to improve the accuracy of dynamic object detection. To further enhance feature stability in complex scenes, the algorithm introduces LSD line features for joint extraction alongside point features. Leveraging the robustness of line features in low-texture or dynamic edge regions, it initially filters out obvious dynamic features. Subsequently, epipolar geometric constraints are applied for a secondary screening of residual dynamic points, achieving high-precision retention of static features. Finally, pose-assisted localization is accomplished using the P3P algorithm based on the filtered static features. Experimental results on the TUM RGB-D dataset demonstrate that the improved algorithm improves localization accuracy by an average of 53.10% compared to ORB-SLAM3, with a maximum improvement of up to 93.56%. On the EuRoC-MAV dataset, the improved algorithm reduces the average ATE-RMSE by 85.32%, with improvements exceeding 85.5% on highly challenging sequences, and maintains an ATE-StD below 0.035m under complex operating conditions, enhancing localization accuracy and anti-interference capability. Physical validation further confirms the superior accuracy and robustness of the proposed method, offering an effective approach for research on high-precision SLAM systems in dynamic environments. ORB-SLAM3 line feature extraction YOLOv8 epipolar geometric constraint P3P pose estimation Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 28 Apr, 2026 Reviews received at journal 27 Apr, 2026 Reviews received at journal 26 Apr, 2026 Reviews received at journal 26 Apr, 2026 Reviewers agreed at journal 23 Apr, 2026 Reviews received at journal 21 Apr, 2026 Reviews received at journal 21 Apr, 2026 Reviewers agreed at journal 21 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 10 Apr, 2026 Reviews received at journal 30 Mar, 2026 Reviewers agreed at journal 30 Mar, 2026 Reviewers invited by journal 26 Mar, 2026 Editor invited by journal 17 Mar, 2026 Editor assigned by journal 16 Mar, 2026 Submission checks completed at journal 11 Mar, 2026 First submitted to journal 11 Mar, 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-8981127","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":613988789,"identity":"48792578-9b2d-421e-9b7e-2b332d5c098a","order_by":0,"name":"jiarui qin","email":"","orcid":"","institution":"Liaoning University of Technology","correspondingAuthor":false,"prefix":"","firstName":"jiarui","middleName":"","lastName":"qin","suffix":""},{"id":613988793,"identity":"435bd499-affa-4817-a548-23f89e8fe660","order_by":1,"name":"yugang wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIiWNgGAWjYJADAwaGCgk5eRK1nLEwNmwgSQtjW0UiwwECyszZzx58XNlmlyfvwLzxwc95EgmMDcwPH93Ao8WyJy/Z8GxbcrHhAbZiw95tEnnsDGzGxjn4HHMgx0yycRtz4sb5b8ykGbdJFDM28LBJ49Vy/o35z8Zt9YkbG3iAWuZIJDYcIKTlRo4ZY+O2w4nzGUBaGojQYjnjXbJk47/jiRsYgH7pOSZhbNhMwC/m/LkHPzacqU6c3wAMsR81dXLy7M0PH+N1GAMPlHEAJsSMRzmKFvkGAipHwSgYBaNg5AIALw5KX1NpLYYAAAAASUVORK5CYII=","orcid":"","institution":"Liaoning University of Technology","correspondingAuthor":true,"prefix":"","firstName":"yugang","middleName":"","lastName":"wang","suffix":""},{"id":613988803,"identity":"143c6c1d-bd84-4527-9ddd-f42f4c37ecc4","order_by":2,"name":"xueli cong","email":"","orcid":"","institution":"Liaoning University of Technology","correspondingAuthor":false,"prefix":"","firstName":"xueli","middleName":"","lastName":"cong","suffix":""},{"id":613988805,"identity":"474c1453-d569-4fb0-bb04-1467bd6b1e41","order_by":3,"name":"liyao zhou","email":"","orcid":"","institution":"Liaoning University of Technology","correspondingAuthor":false,"prefix":"","firstName":"liyao","middleName":"","lastName":"zhou","suffix":""}],"badges":[],"createdAt":"2026-02-26 19:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8981127/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8981127/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105729157,"identity":"3eacb0b2-8f53-4081-89f5-5850ab668863","added_by":"auto","created_at":"2026-03-30 11:13:35","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1239096,"visible":true,"origin":"","legend":"","description":"","filename":"ADeepLearningBasedVisualSLAMApproachforIndoorDynamicScenes.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8981127/v1_covered_172620ee-8ac2-4816-aee8-27b53e9335de.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":" A Deep Learning-Based Visual SLAM Approach for Indoor Dynamic Scenes","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":"discover-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Computing](https://link.springer.com/journal/10791)","snPcode":"10791","submissionUrl":"https://submission.springernature.com/new-submission/10791/3","title":"Discover Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"ORB-SLAM3, line feature extraction, YOLOv8, epipolar geometric constraint, P3P pose estimation","lastPublishedDoi":"10.21203/rs.3.rs-8981127/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8981127/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn indoor dynamic scenes, traditional visual SLAM algorithms often suffer from a significant decline in localization accuracy due to interference from dynamic objects. To address this issue, this study proposes a visual SLAM method based on deep learning. Built upon the ORB-SLAM3 framework, the proposed method effectively enhances the system's capability in complex dynamic environments by constructing a dual dynamic feature suppression mechanism that integrates object detection and epipolar geometric constraints. Specifically, the YOLOv8m network is first lightweighted by introducing GhostConv modules to replace the original convolutional layers, reducing computational overhead. Additionally, the CBAM attention mechanism is incorporated into the backbone network to improve the accuracy of dynamic object detection. To further enhance feature stability in complex scenes, the algorithm introduces LSD line features for joint extraction alongside point features. Leveraging the robustness of line features in low-texture or dynamic edge regions, it initially filters out obvious dynamic features. Subsequently, epipolar geometric constraints are applied for a secondary screening of residual dynamic points, achieving high-precision retention of static features. Finally, pose-assisted localization is accomplished using the P3P algorithm based on the filtered static features. Experimental results on the TUM RGB-D dataset demonstrate that the improved algorithm improves localization accuracy by an average of 53.10% compared to ORB-SLAM3, with a maximum improvement of up to 93.56%. On the EuRoC-MAV dataset, the improved algorithm reduces the average ATE-RMSE by 85.32%, with improvements exceeding 85.5% on highly challenging sequences, and maintains an ATE-StD below 0.035m under complex operating conditions, enhancing localization accuracy and anti-interference capability. Physical validation further confirms the superior accuracy and robustness of the proposed method, offering an effective approach for research on high-precision SLAM systems in dynamic environments.\u003c/p\u003e","manuscriptTitle":" A Deep Learning-Based Visual SLAM Approach for Indoor Dynamic Scenes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-30 06:21:47","doi":"10.21203/rs.3.rs-8981127/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-28T17:31:26+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-27T06:13:51+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-26T08:25:16+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-26T06:43:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"103163122930154888091252234080270151134","date":"2026-04-23T06:12:41+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-21T13:34:49+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-21T12:10:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"147366573345111660960871444549648109830","date":"2026-04-21T04:51:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"55629947535131431781237565425459221056","date":"2026-04-21T01:13:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"216269650366786416634557531120793675056","date":"2026-04-20T21:58:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"60253458093990555298481738393958141330","date":"2026-04-20T21:03:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"172546483546773051996736199495205042408","date":"2026-04-20T19:01:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"43897221474135595154666355035820843816","date":"2026-04-20T16:39:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"130052932397960167468104895316767022348","date":"2026-04-10T08:48:29+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-30T08:24:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"38565226897867048595082609245848643405","date":"2026-03-30T06:21:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-26T11:35:02+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-17T08:15:46+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-16T11:29:06+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-11T17:55:52+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Computing","date":"2026-03-11T11:41:39+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Discover Computing](https://link.springer.com/journal/10791)","snPcode":"10791","submissionUrl":"https://submission.springernature.com/new-submission/10791/3","title":"Discover Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a542fe19-bf62-475e-a251-166841246c93","owner":[],"postedDate":"March 30th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-04-28T17:39:19+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-30 06:21:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8981127","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8981127","identity":"rs-8981127","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}