Efficient Online Tracking-by-Detection With Kalman Filter

Visual tracking of multiple objects in videos has a promisingly broad application in manufacturing, construction, traffic, logistics, etc., especially in large-scale applications where it is not feasible to attach markers to many objects for traditional, marker-enabled tracking methods. This paper p...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Siyuan Chen, Chenhui Shao
Formato: article
Lenguaje:EN
Publicado: IEEE 2021
Materias:
Acceso en línea:https://doaj.org/article/807ee0cd564647518945ab0208c3029e
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:807ee0cd564647518945ab0208c3029e
record_format dspace
spelling oai:doaj.org-article:807ee0cd564647518945ab0208c3029e2021-11-18T00:05:21ZEfficient Online Tracking-by-Detection With Kalman Filter2169-353610.1109/ACCESS.2021.3124705https://doaj.org/article/807ee0cd564647518945ab0208c3029e2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9597516/https://doaj.org/toc/2169-3536Visual tracking of multiple objects in videos has a promisingly broad application in manufacturing, construction, traffic, logistics, etc., especially in large-scale applications where it is not feasible to attach markers to many objects for traditional, marker-enabled tracking methods. This paper presents a new approach, Kalman-intersection-over-union (KIOU) tracker, for multi-object tracking in videos that integrates a Kalman filter with IOU-based track association methods. The performance of the proposed KIOU tracker is quantitatively evaluated with UA-DETRAC, an open real-world multi-object detection and tracking benchmark. Experimental results show that the KIOU tracker outperforms the leading tracking methods. Additionally, the KIOU tracker has speed comparable to simple area overlap-based track association and quality close to methods with much higher computational costs, demonstrating its potential for online, real-time multi-object tracking.Siyuan ChenChenhui ShaoIEEEarticleComputer visionKalman filtermulti-object trackingtracking-by-detectiononline trackingtransportationElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 147570-147578 (2021)
institution DOAJ
collection DOAJ
language EN
topic Computer vision
Kalman filter
multi-object tracking
tracking-by-detection
online tracking
transportation
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
spellingShingle Computer vision
Kalman filter
multi-object tracking
tracking-by-detection
online tracking
transportation
Electrical engineering. Electronics. Nuclear engineering
TK1-9971
Siyuan Chen
Chenhui Shao
Efficient Online Tracking-by-Detection With Kalman Filter
description Visual tracking of multiple objects in videos has a promisingly broad application in manufacturing, construction, traffic, logistics, etc., especially in large-scale applications where it is not feasible to attach markers to many objects for traditional, marker-enabled tracking methods. This paper presents a new approach, Kalman-intersection-over-union (KIOU) tracker, for multi-object tracking in videos that integrates a Kalman filter with IOU-based track association methods. The performance of the proposed KIOU tracker is quantitatively evaluated with UA-DETRAC, an open real-world multi-object detection and tracking benchmark. Experimental results show that the KIOU tracker outperforms the leading tracking methods. Additionally, the KIOU tracker has speed comparable to simple area overlap-based track association and quality close to methods with much higher computational costs, demonstrating its potential for online, real-time multi-object tracking.
format article
author Siyuan Chen
Chenhui Shao
author_facet Siyuan Chen
Chenhui Shao
author_sort Siyuan Chen
title Efficient Online Tracking-by-Detection With Kalman Filter
title_short Efficient Online Tracking-by-Detection With Kalman Filter
title_full Efficient Online Tracking-by-Detection With Kalman Filter
title_fullStr Efficient Online Tracking-by-Detection With Kalman Filter
title_full_unstemmed Efficient Online Tracking-by-Detection With Kalman Filter
title_sort efficient online tracking-by-detection with kalman filter
publisher IEEE
publishDate 2021
url https://doaj.org/article/807ee0cd564647518945ab0208c3029e
work_keys_str_mv AT siyuanchen efficientonlinetrackingbydetectionwithkalmanfilter
AT chenhuishao efficientonlinetrackingbydetectionwithkalmanfilter
_version_ 1718425233268932608