An Anthropometric Measurement Method for Forensic Images Based on Deep Keypoint Recognition and Dynamic Calibration Fusion | 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 An Anthropometric Measurement Method for Forensic Images Based on Deep Keypoint Recognition and Dynamic Calibration Fusion Bingzhe Wang, Liping Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9121260/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract To improve the efficiency, accuracy, and scene adaptability of anthropometric measurement from forensic images, this study proposes a height-measurement method integrating deep keypoint recognition with dynamic calibration. YOLOv11-Pose detects 17 human keypoints, while a cross-shaped electric telescopic rig and an Intel RealSense D435 RGB-D camera provide real-time pixel-to-physical scale mapping for automatic height estimation. Experiments under typical surveillance distances and viewing angles achieved a mean absolute error (MAE) of 0.65–1.98 cm and a P95 of 0.77–2.44 cm. Error increased with distance, and tail risk became more pronounced at larger viewing angles. In ablation tests, multi-marker weighted fusion reduced MAE and P95 by 75.7% and 72.6%, respectively, relative to a single-marker baseline; under challenging conditions, the reductions remained 50.4% and 61.9%. The method supports sub-second single-frame processing while improving scale-recovery robustness and suppressing long-tail errors. It is suitable for suspect identification, crime-scene reconstruction, and quantitative evidential analysis from forensic imagery. forensic images anthropometric measurement method deep keypoint recognition dynamic calibration multi-marker fusion height measurement Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 17 Mar, 2026 Editor assigned by journal 16 Mar, 2026 Submission checks completed at journal 16 Mar, 2026 First submitted to journal 14 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-9121260","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":607337582,"identity":"749efb29-e2f6-4a81-873c-e80e012e5149","order_by":0,"name":"Bingzhe Wang","email":"","orcid":"","institution":"People's Public Security University of China","correspondingAuthor":false,"prefix":"","firstName":"Bingzhe","middleName":"","lastName":"Wang","suffix":""},{"id":607337584,"identity":"eb8617b3-8370-41be-87e0-6133f46a751a","order_by":1,"name":"Liping Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYDACZsaGAx8M2Or5mZkPPgDyefgIamFnPnhwRgVfgmR7W7IBSAsbQS38bMmHOc7IJRj0nFGTAAkQ1GJwmMfgMGObWZ6BRA5b5dccOxk2BuaHj24Q0lLYllZsLpF77LbstmSgw9iMjXPwaDEDaZnZdoxx54y8tNuS25iBWnjYpAlq4W37z7jhRo5ZseS2emK0sCUc5jnDlrjhzBkzxo/bDhPWYn+Y+QAwkNmMQYEszbjtOA8bMwG/SPYfbP4AjEo5UFR+/Lmt2p6fvfnhY3xaUAAzD5gkVjkIMP4gRfUoGAWjYBSMGAAAKd9MSfCJNPQAAAAASUVORK5CYII=","orcid":"","institution":"People's Public Security University of China","correspondingAuthor":true,"prefix":"","firstName":"Liping","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2026-03-14 09:23:39","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9121260/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9121260/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108181940,"identity":"24f6d24f-3e4d-46da-94ca-a7f319137dbe","added_by":"auto","created_at":"2026-04-30 08:59:02","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2643216,"visible":true,"origin":"","legend":"","description":"","filename":"AnAnthropometricMeasurementMethodforForensicImagesBasedonDeepKeypointRecognitionandDynamicCalibrationFusion.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9121260/v1_covered_3677a187-0b2e-48ba-910c-39004ac26263.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"An Anthropometric Measurement Method for Forensic Images Based on Deep Keypoint Recognition and Dynamic Calibration Fusion","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":"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":"forensic images, anthropometric measurement method, deep keypoint recognition, dynamic calibration, multi-marker fusion, height measurement","lastPublishedDoi":"10.21203/rs.3.rs-9121260/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9121260/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eTo improve the efficiency, accuracy, and scene adaptability of anthropometric measurement from forensic images, this study proposes a height-measurement method integrating deep keypoint recognition with dynamic calibration. YOLOv11-Pose detects 17 human keypoints, while a cross-shaped electric telescopic rig and an Intel RealSense D435 RGB-D camera provide real-time pixel-to-physical scale mapping for automatic height estimation. Experiments under typical surveillance distances and viewing angles achieved a mean absolute error (MAE) of 0.65\u0026ndash;1.98 cm and a P95 of 0.77\u0026ndash;2.44 cm. Error increased with distance, and tail risk became more pronounced at larger viewing angles. In ablation tests, multi-marker weighted fusion reduced MAE and P95 by 75.7% and 72.6%, respectively, relative to a single-marker baseline; under challenging conditions, the reductions remained 50.4% and 61.9%. The method supports sub-second single-frame processing while improving scale-recovery robustness and suppressing long-tail errors. It is suitable for suspect identification, crime-scene reconstruction, and quantitative evidential analysis from forensic imagery. \u003c/p\u003e","manuscriptTitle":"An Anthropometric Measurement Method for Forensic Images Based on Deep Keypoint Recognition and Dynamic Calibration Fusion","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-29 09:57:18","doi":"10.21203/rs.3.rs-9121260/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-17T05:38:12+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-17T01:39:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-17T01:39:34+00:00","index":"","fulltext":""},{"type":"submitted","content":"Signal, Image and Video Processing","date":"2026-03-14T09:10:27+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":"5bf6ae97-88e4-4e6a-a125-5c3f1803d01d","owner":[],"postedDate":"April 29th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-29T09:57:18+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-29 09:57:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9121260","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9121260","identity":"rs-9121260","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.