Rapid Ghost Pedestrian Detection: Spike Vision for Safer Autonomous Driving

Rapid Ghost Pedestrian Detection: Spike Vision for Safer Autonomous Driving | 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 Article Rapid Ghost Pedestrian Detection: Spike Vision for Safer Autonomous Driving Yajing Zheng, Tiejun Huang, Baoyue Zhang, Zhaofei Yu, Jiyuan Zhang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5491257/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Autonomous driving and robotics are increasingly focusing on improving safety, particularly in collision avoidance systems. Detecting unexpected obstacles quickly is essential for preventing accidents, especially in high-speed driving scenarios where the sudden appearance of pedestrians presents a significant challenge. Conventional systems often rely on prediction, but they struggle with obstacles emerging suddenly from blind spots, introducing delays in detection. Here, we show that spike cameras, can significantly reduce these delays by enabling rapid object detection. Our system integrates perception, reasoning, and action into an end-to-end, ultra-low-latency closed-loop that allows for faster reactions. Our system reduces the overall latency of detecting sudden appearance of objects to under 5 ms: in real-world ghosting pedestrian scenarios, it achieves a 97% success rate in obstacle avoidance tests. Additionally, even at an equivalent real vehicle speed of 118 km/h, the system maintains a success rate of 92.5%. These results represent a meaningful improvement in autonomous driving safety, providing faster, more accurate obstacle detection in high-speed environments, where quick reactions are crucial for avoiding accidents and ensuring the safety of all road users. Physical sciences/Mathematics and computing/Computational science Physical sciences/Engineering Autonomous driving safety Collision avoidance Sudden obstacle detection Neuromorphic vision sensors Rapid perception systems Spike cameras Full Text Additional Declarations There is NO Competing Interest. Supplementary Files supplementaryvideo.mp4 Ultra-Low-Latency Ghost Pedestrian Detection with Spike Vision Cite Share Download PDF Status: Under Review 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-5491257","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":479283684,"identity":"c2491891-e46b-4ffa-ba4f-292c916d297f","order_by":0,"name":"Yajing Zheng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIie3OMQuCQBTA8SdCLWeu1xD1EQwhAkG/inKQS0RjQ8NBcGtr0McImk8e3GS7kHNz4C550ay2Bd0f3r3l/eAATKafTIKtlwsDvewvyJj3J58zT/YlXibtarsrw8tdrCjsgoQPb7KdSAn+KX+wa6kUhTxNONnEnYQ5AtmiSAW1BCacEq+ToFMj80+a1P2IdXA4hh5dKWrxHmQss4NNFMa0VGwZq9QXZN1ORgViRfYYuWcxL577YHIc5u1kVmRc7+Y/zRs3M2i9b5oe3wIioF2nJpPJ9Le9AFOsStGfoMzbAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-6355-7354","institution":"Peking University","correspondingAuthor":true,"prefix":"","firstName":"Yajing","middleName":"","lastName":"Zheng","suffix":""},{"id":479283685,"identity":"a9abcf64-29c4-4061-86e3-979121cf7eba","order_by":1,"name":"Tiejun Huang","email":"","orcid":"https://orcid.org/0000-0002-4234-6099","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Tiejun","middleName":"","lastName":"Huang","suffix":""},{"id":479283686,"identity":"6d419e2b-b7fb-4a01-b019-cadd2afd9361","order_by":2,"name":"Baoyue Zhang","email":"","orcid":"","institution":"Peking University, Beijing","correspondingAuthor":false,"prefix":"","firstName":"Baoyue","middleName":"","lastName":"Zhang","suffix":""},{"id":479283687,"identity":"e220793e-8a98-468b-97d9-173ac34e21fe","order_by":3,"name":"Zhaofei Yu","email":"","orcid":"https://orcid.org/0000-0002-6913-7553","institution":"Peking University, Beijing","correspondingAuthor":false,"prefix":"","firstName":"Zhaofei","middleName":"","lastName":"Yu","suffix":""},{"id":479283688,"identity":"7cd13b80-02e4-4005-a2dd-bbb594e01f5d","order_by":4,"name":"Jiyuan Zhang","email":"","orcid":"","institution":"Peking University","correspondingAuthor":false,"prefix":"","firstName":"Jiyuan","middleName":"","lastName":"Zhang","suffix":""},{"id":479283689,"identity":"be8022d8-374e-4de4-8526-760d3c06f2bc","order_by":5,"name":"Jinwei Chen","email":"","orcid":"","institution":"Spikesee (Beijing) Technology Co. Ltd.","correspondingAuthor":false,"prefix":"","firstName":"Jinwei","middleName":"","lastName":"Chen","suffix":""},{"id":479283690,"identity":"b16c6417-9650-41fc-bd8c-bf12efa5a852","order_by":6,"name":"Karl Friston","email":"","orcid":"https://orcid.org/0000-0001-7984-8909","institution":"University College London","correspondingAuthor":false,"prefix":"","firstName":"Karl","middleName":"","lastName":"Friston","suffix":""}],"badges":[],"createdAt":"2024-11-20 13:25:51","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5491257/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5491257/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":87368219,"identity":"e6a0f66d-3d6c-44d1-a908-ce1c6cdbe7a5","added_by":"auto","created_at":"2025-07-23 06:52:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1301680,"visible":true,"origin":"","legend":"Article File","description":"","filename":"manuscriptspikeAuto.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5491257/v1_covered_668c293d-0612-45e5-93b9-987e95191f0e.pdf"},{"id":87367215,"identity":"f817384a-cda4-4c2a-a120-5a61667d8509","added_by":"auto","created_at":"2025-07-23 06:44:38","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":84720552,"visible":true,"origin":"","legend":"Ultra-Low-Latency Ghost Pedestrian Detection with Spike Vision","description":"","filename":"supplementaryvideo.mp4","url":"https://assets-eu.researchsquare.com/files/rs-5491257/v1/0e9d5b893d5d323d896a46e5.mp4"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Rapid Ghost Pedestrian Detection: Spike Vision for Safer Autonomous Driving","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":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Autonomous driving safety, Collision avoidance, Sudden obstacle detection, Neuromorphic vision sensors, Rapid perception systems, Spike cameras","lastPublishedDoi":"10.21203/rs.3.rs-5491257/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5491257/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Autonomous driving and robotics are increasingly focusing on improving safety, particularly in collision avoidance systems. Detecting unexpected obstacles quickly is essential for preventing accidents, especially in high-speed driving scenarios where the sudden appearance of pedestrians presents a significant challenge. Conventional systems often rely on prediction, but they struggle with obstacles emerging suddenly from blind spots, introducing delays in detection. Here, we show that spike cameras, can significantly reduce these delays by enabling rapid object detection. Our system integrates perception, reasoning, and action into an end-to-end, ultra-low-latency closed-loop that allows for faster reactions. Our system reduces the overall latency of detecting sudden appearance of objects to under 5 ms: in real-world ghosting pedestrian scenarios, it achieves a 97% success rate in obstacle avoidance tests. Additionally, even at an equivalent real vehicle speed of 118 km/h, the system maintains a success rate of 92.5%. These results represent a meaningful improvement in autonomous driving safety, providing faster, more accurate obstacle detection in high-speed environments, where quick reactions are crucial for avoiding accidents and ensuring the safety of all road users.","manuscriptTitle":"Rapid Ghost Pedestrian Detection: Spike Vision for Safer Autonomous Driving","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-23 06:44:25","doi":"10.21203/rs.3.rs-5491257/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"communications-engineering","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"commseng","sideBox":"Learn more about [Communications Engineering](http://link.springer.com/journal/44172)","snPcode":"44172","submissionUrl":"https://mts-commseng.nature.com/cgi-bin/main.plex","title":"Communications Engineering","twitterHandle":"@commseng","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"64df416e-b60d-45cd-ba5a-0f86c4149aec","owner":[],"postedDate":"July 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":50891199,"name":"Physical sciences/Mathematics and computing/Computational science"},{"id":50891200,"name":"Physical sciences/Engineering"}],"tags":[],"updatedAt":"2025-07-23T06:44:25+00:00","versionOfRecord":[],"versionCreatedAt":"2025-07-23 06:44:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5491257","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5491257","identity":"rs-5491257","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}