QueryTrack: identifying and tracking a person of interest using clothing-based hybrid features

Basit Öğe Kaydı
dc.contributor.authorOrtac kosun, Gizem
dc.contributor.authorYilmaz, Seckin
dc.contributor.authorSamli, Ruya
dc.date.accessioned2026-02-12T21:05:43Z
dc.date.available2026-02-12T21:05:43Z
dc.date.issued2026
dc.departmentBursa Teknik Üniversitesi
dc.description.abstractLocating and tracking a specific person of interest in a single visual query remains a significant challenge in complex surveillance environments. Current paradigms fall short: generic multi-object trackers suffer from identity loss over time, while existing person search methods, designed for static image galleries, lack robustness against the dynamic complexities of video streams, especially occlusions. This paper introduces QueryTrack, a comprehensive framework designed specifically for this query-based tracking task. The core novelty lies in a powerful re-identification engine that fuses four distinct feature types-HOG, Gabor, Color, and VGG16-into a highly discriminative signature for the target. This signature drives a hybrid tracking algorithm that synergizes motion prediction and visual tracking to maintain identity continuity. Furthermore, we propose a new post-occlusion recovery technique to handle long-term disappearances. Experimental evaluations validate our method's superior performance, achieving F1 scores of 97.20%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$97.20\%$$\end{document} in crowded scenarios and 96.35%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$96.35\%$$\end{document} with minimal occlusion, confirming its significant contribution to accurate and persistent person tracking under realistic conditions. Additionally, we provide a transparent computational cost analysis, confirming the system's viability for offline forensic investigation where accuracy is paramount.
dc.identifier.doi10.1007/s00371-025-04339-0
dc.identifier.issn0178-2789
dc.identifier.issn1432-2315
dc.identifier.issue3
dc.identifier.scopus2-s2.0-105028953713
dc.identifier.scopusqualityQ1
dc.identifier.urihttps://doi.org/10.1007/s00371-025-04339-0
dc.identifier.urihttps://hdl.handle.net/20.500.12885/7113
dc.identifier.volume42
dc.identifier.wosWOS:001675332400001
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.language.isoen
dc.publisherSpringer
dc.relation.ispartofVisual Computer
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.snmzKA_WoS_20260212
dc.subjectVideo surveillance
dc.subjectPerson re-identification
dc.subjectPerson tracking
dc.subjectMulti-person tracking
dc.subjectSoft biometry
dc.subjectClothing-based identification
dc.titleQueryTrack: identifying and tracking a person of interest using clothing-based hybrid features
dc.typeReview Article

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu