GlyphMotion - A Low-Latency High-Fidelity 4K YOLOv8- Based Multi-Object Tracking Pipeline with Asynchronous Architecture, Compression-Aware Signal Conditioning, and High-Frequency Detail Reinjection

preprint OA: closed CC-BY-4.0
AI-generated deep summary by claude@2026-06, 2026-06-24 · read from full text

GlyphMotion studies the performance of a 4K multi-object tracking pipeline that uses YOLOv8 with ByteTrack, focusing on reducing latency jitter and maintaining audio-visual synchronization during video processing. The authors propose a three-thread asynchronous architecture using FFmpeg subprocesses plus High-Frequency Detail Reinjection (HFDR), and report a 92.4% latency jitter reduction on modern hardware and a 99.8% reduction on legacy hardware; they also find that CRF 24 compression improves MOTA from 55.36 to 85.58 by stabilizing identity associations, while HFDR attains a VMAF score of 96.72. A major caveat is that the work is presented as a preprint with no journal peer review, though it is evaluated across a 420-experiment controlled benchmark and validated on two GPU generations with an automated pre-flight codec analysis gate (recon.py) to ensure measurement validity. The paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

Abstract Synchronous multi-object tracking pipelines processing 4K video suffer from catastrophic latency jitter, blocking input/output operations, native audio loss, and tracking degradation due to raw sensor noise. GlyphMotion introduces a novel three-thread asynchronous architecture coupled with FFmpeg subprocesses and High-Frequency Detail Reinjection (HFDR) to decouple inference from encoding. This decoupling achieves a 92.4% reduction in latency jitter on modern hardware (337.8 ms to 27.04 ms) and a 99.8% reduction on legacy hardware (21,342 ms to 46.5 ms). Furthermore, this paper identifies that Constant Rate Factor (CRF) 24 compression acts as a highly effective signal conditioner, surprisingly improving YOLOv8 Multi-Object Tracking Accuracy (MOTA) from a raw baseline of 55.36 to an unprecedented production level of 85.58 by stabilising ByteTrack identity associations. While compression aids tracking, the HFDR engine simultaneously recovers perceptual quality, achieving a Video Multi-Method Assessment Fusion (VMAF) score of 96.72. Evaluated across a 420-experiment controlled benchmark and validated on two distinct GPU generations, the production NVENC configuration natively preserves audio synchronisation and outperforms existing synchronous baselines while establishing that pipeline architecture is equally as critical as underlying model architecture. Furthermore, the pipeline integrates an automated pre-flight codec analysis gate (recon.py) that enforces input conformance, validating a VMAF recovery from 0.963 to 97.186 on non-conformant mobile phone footage and ensuring measurement validity across all reported quality scores.
Full text 12,285 characters · extracted from preprint-html · click to expand
GlyphMotion - A Low-Latency High-Fidelity 4K YOLOv8- Based Multi-Object Tracking Pipeline with Asynchronous Architecture, Compression-Aware Signal Conditioning, and High-Frequency Detail Reinjection | 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 GlyphMotion - A Low-Latency High-Fidelity 4K YOLOv8- Based Multi-Object Tracking Pipeline with Asynchronous Architecture, Compression-Aware Signal Conditioning, and High-Frequency Detail Reinjection Shitij Halder, Sayan Sarkar, Dr. Kretika Goel This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9552341/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 Synchronous multi-object tracking pipelines processing 4K video suffer from catastrophic latency jitter, blocking input/output operations, native audio loss, and tracking degradation due to raw sensor noise. GlyphMotion introduces a novel three-thread asynchronous architecture coupled with FFmpeg subprocesses and High-Frequency Detail Reinjection (HFDR) to decouple inference from encoding. This decoupling achieves a 92.4% reduction in latency jitter on modern hardware (337.8 ms to 27.04 ms) and a 99.8% reduction on legacy hardware (21,342 ms to 46.5 ms). Furthermore, this paper identifies that Constant Rate Factor (CRF) 24 compression acts as a highly effective signal conditioner, surprisingly improving YOLOv8 Multi-Object Tracking Accuracy (MOTA) from a raw baseline of 55.36 to an unprecedented production level of 85.58 by stabilising ByteTrack identity associations. While compression aids tracking, the HFDR engine simultaneously recovers perceptual quality, achieving a Video Multi-Method Assessment Fusion (VMAF) score of 96.72. Evaluated across a 420-experiment controlled benchmark and validated on two distinct GPU generations, the production NVENC configuration natively preserves audio synchronisation and outperforms existing synchronous baselines while establishing that pipeline architecture is equally as critical as underlying model architecture. Furthermore, the pipeline integrates an automated pre-flight codec analysis gate (recon.py) that enforces input conformance, validating a VMAF recovery from 0.963 to 97.186 on non-conformant mobile phone footage and ensuring measurement validity across all reported quality scores. Asynchronous Pipeline Video Processing Optimisation High-Frequency Detail Reinjection Latency Jitter Reduction Audio-Visual Synchronisation Multi-Object Tracking 4K Video Analytics HEVC Hardware Encoding Cross-Hardware Validation ByteTrack YOLOv8 Input Video Normalisation Full Text Additional Declarations No competing interests reported. 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-9552341","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":630960651,"identity":"4ddb868a-38a1-4b85-85f3-41d4d1fa63a5","order_by":0,"name":"Shitij Halder","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAElEQVRIiWNgGAWjYDACCcYGhgSeAwx8DAlATgVQhJm5gZCWxgaQFjaQFoszIC2MhLQwgFRAtVS2gYQIaJGf3dz+4IHMHXk29uRjH27Oq43mbwdq+VGxDacWgzsHQQ57ZtjG8yx55sxtx3NnHGZsYOw5cxu3FolEkJbDjG0SOcbMktuO5TYAtTAztuHWIj8DosW+TSL/M/PfOcdy5xPSwnADoiURaAszg2RDTe4GQloMgFpmALUkA/1izCBx7EDuRqCWg/j8Ij8j/cHHnz2HbfvZkx8zSNTU5c47f/jggx8VeBwGAow9cOZhMHkAv3oQ+AFn1RFWPApGwSgYBSMOAACZtmIb7KedagAAAABJRU5ErkJggg==","orcid":"","institution":"Dehradun Institute of Technology University","correspondingAuthor":true,"prefix":"","firstName":"Shitij","middleName":"","lastName":"Halder","suffix":""},{"id":630960652,"identity":"3b133335-d6b6-4522-bc0a-e1eae1ac890e","order_by":1,"name":"Sayan Sarkar","email":"","orcid":"","institution":"Singur General Degree College","correspondingAuthor":false,"prefix":"","firstName":"Sayan","middleName":"","lastName":"Sarkar","suffix":""},{"id":630960653,"identity":"4742be3d-d726-4efa-a7d4-7dbd085d28b1","order_by":2,"name":"Dr. Kretika Goel","email":"","orcid":"","institution":"Dehradun Institute of Technology University","correspondingAuthor":false,"prefix":"Dr.","firstName":"Kretika","middleName":"","lastName":"Goel","suffix":""}],"badges":[],"createdAt":"2026-04-28 10:10:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9552341/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9552341/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108492738,"identity":"19d0316d-0bc4-47a4-afad-88b0fd6232e7","added_by":"auto","created_at":"2026-05-05 09:58:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1556855,"visible":true,"origin":"","legend":"","description":"","filename":"GlyphMotionALowLatencyHighFidelity4KYOLOv8BasedMultiObjectTrackingPipelinewithAsynchronousArchitectureCompressionAwareSignalConditioningandHighFrequencyDetailReinjection.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9552341/v1_covered_d139d7ff-499e-4427-910e-198fef8c8765.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"GlyphMotion - A Low-Latency High-Fidelity 4K YOLOv8- Based Multi-Object Tracking Pipeline with Asynchronous Architecture, Compression-Aware Signal Conditioning, and High-Frequency Detail Reinjection","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"Asynchronous Pipeline, Video Processing Optimisation, High-Frequency Detail Reinjection, Latency Jitter Reduction, Audio-Visual Synchronisation, Multi-Object Tracking, 4K Video Analytics, HEVC Hardware Encoding, Cross-Hardware Validation, ByteTrack, YOLOv8, Input Video Normalisation","lastPublishedDoi":"10.21203/rs.3.rs-9552341/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9552341/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSynchronous multi-object tracking pipelines processing 4K video suffer from catastrophic latency jitter, blocking input/output operations, native audio loss, and tracking degradation due to raw sensor noise. GlyphMotion introduces a novel three-thread asynchronous architecture coupled with FFmpeg subprocesses and High-Frequency Detail Reinjection (HFDR) to decouple inference from encoding. This decoupling achieves a 92.4% reduction in latency jitter on modern hardware (337.8 ms to 27.04 ms) and a 99.8% reduction on legacy hardware (21,342 ms to 46.5 ms). Furthermore, this paper identifies that Constant Rate Factor (CRF) 24 compression acts as a highly effective signal conditioner, surprisingly improving YOLOv8 Multi-Object Tracking Accuracy (MOTA) from a raw baseline of 55.36 to an unprecedented production level of 85.58 by stabilising ByteTrack identity associations. While compression aids tracking, the HFDR engine simultaneously recovers perceptual quality, achieving a Video Multi-Method Assessment Fusion (VMAF) score of 96.72. Evaluated across a 420-experiment controlled benchmark and validated on two distinct GPU generations, the production NVENC configuration natively preserves audio synchronisation and outperforms existing synchronous baselines while establishing that pipeline architecture is equally as critical as underlying model architecture. Furthermore, the pipeline integrates an automated pre-flight codec analysis gate (recon.py) that enforces input conformance, validating a VMAF recovery from 0.963 to 97.186 on non-conformant mobile phone footage and ensuring measurement validity across all reported quality scores.\u003c/p\u003e","manuscriptTitle":"GlyphMotion - A Low-Latency High-Fidelity 4K YOLOv8- Based Multi-Object Tracking Pipeline with Asynchronous Architecture, Compression-Aware Signal Conditioning, and High-Frequency Detail Reinjection","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-29 05:51:14","doi":"10.21203/rs.3.rs-9552341/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":"c6473671-d61b-400e-8253-92a9721b2a01","owner":[],"postedDate":"April 29th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Rejected","date":"2026-05-04T12:26:00+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-05-04T12:24:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-05-04T06:09:45+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-04T18:14:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-29 05:51:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9552341","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9552341","identity":"rs-9552341","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.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2026) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-06-05T02:00:03.366016+00:00
License: CC-BY-4.0