PFSAN: An Efficient Perception Network for Autonomous Driving Based on Progressive Feature Splitting and Aggregation

PFSAN: An Efficient Perception Network for Autonomous Driving Based on Progressive Feature Splitting and 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 PFSAN: An Efficient Perception Network for Autonomous Driving Based on Progressive Feature Splitting and Aggregation Yang Cui, Yi Han, Dong Guo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6208099/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 30 Jul, 2025 Read the published version in Signal, Image and Video Processing → Version 1 posted 5 You are reading this latest preprint version Abstract Bottleneck structures are widely used in object detection networks but often face challenges related to limited feature representation and computational efficiency, especially in lightweight designs. To address this, this paper introduces a novel Progressive Feature Splitting and Aggregation Module (PFSAM) that improves feature extraction through feature decomposition and aggregation. It also integrates structural re-parameterization to reduce computational redundancy and optimize gradient flow. Using PFSAM, several network variants are designed to meet varying computational resource requirements. Experimental results demonstrate that the PFSAM module achieves the fastest inference speed and exceptional efficiency with minimal computational complexity and parameter count. In the BDD100K dataset, the PFSAN network excels in small object detection and complex background scenarios, striking a balance between lightweight design and high performance. It provides an efficient solution for object detection tasks. Object Detection Deep learning Feature fusion Autonomous driving Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 30 Jul, 2025 Read the published version in Signal, Image and Video Processing → Version 1 posted Editorial decision: Revision requested 22 Mar, 2025 Reviewers invited by journal 22 Mar, 2025 Editor assigned by journal 13 Mar, 2025 Submission checks completed at journal 13 Mar, 2025 First submitted to journal 11 Mar, 2025 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-6208099","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":432571227,"identity":"cf6bcc49-f43e-4ac5-9a5c-7112f75854f4","order_by":0,"name":"Yang Cui","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIie3NsUvEMBTH8VcCccm16zsKvX8hR8DpwH/lBeGmDjcdbgqFOIlr/S8EwTlHoS7x5o7t4qx0clFzs0d6bg75QjI8+PADiMX+YRlj1tLVSvCssgy4P9kJMr81uu/dukixpdOIdE4tB9OoAkp5GoGOzpE40wbcx7jZQpF2lIybgEhqWiMJrg27e8rrPah5RyyvA4QhtUgo/MrrM5sZ0I+dHxUBwlEbJOl/KN8O5HqSCNEwSSQVh5IfCMkpgmcm6clSwbFVudjj8sENVR4iF032vvv8+haL+2oYxXa1SF8ud2OI/Fr1L7n5A4jFYrHYsX4AGThLmPQwO88AAAAASUVORK5CYII=","orcid":"","institution":"Chang'an University","correspondingAuthor":true,"prefix":"","firstName":"Yang","middleName":"","lastName":"Cui","suffix":""},{"id":432571230,"identity":"9935070b-679e-4b23-8b65-0dff5d9ea39a","order_by":1,"name":"Yi Han","email":"","orcid":"","institution":"Chang'an University","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Han","suffix":""},{"id":432571233,"identity":"984ce28d-1e8a-489f-ab55-b2e92bc0349e","order_by":2,"name":"Dong Guo","email":"","orcid":"","institution":"Chang'an University","correspondingAuthor":false,"prefix":"","firstName":"Dong","middleName":"","lastName":"Guo","suffix":""}],"badges":[],"createdAt":"2025-03-12 03:23:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6208099/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6208099/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11760-025-04551-4","type":"published","date":"2025-07-30T16:38:16+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":88268491,"identity":"607da5b9-b648-414c-bcbc-d65ade9aa366","added_by":"auto","created_at":"2025-08-04 16:52:14","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2456077,"visible":true,"origin":"","legend":"","description":"","filename":"PFSAN.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6208099/v1_covered_4a610d07-6910-416b-a207-698e47c3675f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"PFSAN: An Efficient Perception Network for Autonomous Driving Based on Progressive Feature Splitting and Aggregation","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":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Object Detection, Deep learning, Feature fusion, Autonomous driving","lastPublishedDoi":"10.21203/rs.3.rs-6208099/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6208099/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Bottleneck structures are widely used in object detection networks but often face challenges related to limited feature representation and computational efficiency, especially in lightweight designs. To address this, this paper introduces a novel Progressive Feature Splitting and Aggregation Module (PFSAM) that improves feature extraction through feature decomposition and aggregation. It also integrates structural re-parameterization to reduce computational redundancy and optimize gradient flow. Using PFSAM, several network variants are designed to meet varying computational resource requirements. Experimental results demonstrate that the PFSAM module achieves the fastest inference speed and exceptional efficiency with minimal computational complexity and parameter count. In the BDD100K dataset, the PFSAN network excels in small object detection and complex background scenarios, striking a balance between lightweight design and high performance. It provides an efficient solution for object detection tasks.","manuscriptTitle":"PFSAN: An Efficient Perception Network for Autonomous Driving Based on Progressive Feature Splitting and Aggregation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-16 04:03:16","doi":"10.21203/rs.3.rs-6208099/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-03-22T22:29:34+00:00","index":"","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-03-22T22:28:57+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-03-13T12:46:12+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-03-13T12:43:05+00:00","index":"","fulltext":""},{"type":"submitted","content":"Signal, Image and Video Processing","date":"2025-03-12T03:08:47+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"b90daa87-692e-4972-88e5-adbfb739fea0","owner":[],"postedDate":"April 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-08-04T16:47:04+00:00","versionOfRecord":{"articleIdentity":"rs-6208099","link":"https://doi.org/10.1007/s11760-025-04551-4","journal":{"identity":"signal-image-and-video-processing","isVorOnly":false,"title":"Signal, Image and Video Processing"},"publishedOn":"2025-07-30 16:38:16","publishedOnDateReadable":"July 30th, 2025"},"versionCreatedAt":"2025-04-16 04:03:16","video":"","vorDoi":"10.1007/s11760-025-04551-4","vorDoiUrl":"https://doi.org/10.1007/s11760-025-04551-4","workflowStages":[]},"version":"v1","identity":"rs-6208099","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6208099","identity":"rs-6208099","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}