Multiple object tracking in unknown backgrounds with labeled random finite sets

This paper proposes an online multiple object tracker that can operate under unknown detection profile and clutter rate. In a majority of multiple object tracking applications, model parameters for background processes such as clutter and detection are unknown and vary with time; hence, the ability...

Full description

Bibliographic Details
Main Authors: Punchihewa, Y., Vo, Ba Tuong, Vo, B., Kim, Du Yong
Format: Journal Article
Published: IEEE 2018
Online Access:http://purl.org/au-research/grants/arc/DP160104662
http://hdl.handle.net/20.500.11937/66760
_version_ 1848761386061004800
author Punchihewa, Y.
Vo, Ba Tuong
Vo, B.
Kim, Du Yong
author_facet Punchihewa, Y.
Vo, Ba Tuong
Vo, B.
Kim, Du Yong
author_sort Punchihewa, Y.
building Curtin Institutional Repository
collection Online Access
description This paper proposes an online multiple object tracker that can operate under unknown detection profile and clutter rate. In a majority of multiple object tracking applications, model parameters for background processes such as clutter and detection are unknown and vary with time; hence, the ability of the algorithm to adaptively learn these parameters is essential in practice. In this paper, we detail how the generalized labeled multibernoulli filter, a tractable and provably Bayes optimal multiobject tracker, can be tailored to learn clutter and detection parameters on-the-fly while tracking. Provided that these background model parameters do not fluctuate rapidly compared to the data rate, the proposed algorithm can adapt to the unknown background yielding better tracking performance.
first_indexed 2025-11-14T10:30:50Z
format Journal Article
id curtin-20.500.11937-66760
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T10:30:50Z
publishDate 2018
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-667602022-10-27T06:25:05Z Multiple object tracking in unknown backgrounds with labeled random finite sets Punchihewa, Y. Vo, Ba Tuong Vo, B. Kim, Du Yong This paper proposes an online multiple object tracker that can operate under unknown detection profile and clutter rate. In a majority of multiple object tracking applications, model parameters for background processes such as clutter and detection are unknown and vary with time; hence, the ability of the algorithm to adaptively learn these parameters is essential in practice. In this paper, we detail how the generalized labeled multibernoulli filter, a tractable and provably Bayes optimal multiobject tracker, can be tailored to learn clutter and detection parameters on-the-fly while tracking. Provided that these background model parameters do not fluctuate rapidly compared to the data rate, the proposed algorithm can adapt to the unknown background yielding better tracking performance. 2018 Journal Article http://hdl.handle.net/20.500.11937/66760 10.1109/TSP.2018.2821650 http://purl.org/au-research/grants/arc/DP160104662 IEEE restricted
spellingShingle Punchihewa, Y.
Vo, Ba Tuong
Vo, B.
Kim, Du Yong
Multiple object tracking in unknown backgrounds with labeled random finite sets
title Multiple object tracking in unknown backgrounds with labeled random finite sets
title_full Multiple object tracking in unknown backgrounds with labeled random finite sets
title_fullStr Multiple object tracking in unknown backgrounds with labeled random finite sets
title_full_unstemmed Multiple object tracking in unknown backgrounds with labeled random finite sets
title_short Multiple object tracking in unknown backgrounds with labeled random finite sets
title_sort multiple object tracking in unknown backgrounds with labeled random finite sets
url http://purl.org/au-research/grants/arc/DP160104662
http://hdl.handle.net/20.500.11937/66760