OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance

© 2017 IEEE. The optimal sub-pattern assignment (OSPA) metric is a distance between two sets of points that jointly accounts for the dissimilarity in the number of points and the values of the points in the respective sets. The OSPA metric is often used for measuring the distance between two sets of...

Full description

Bibliographic Details
Main Authors: Beard, Michael, Vo, Ba Tuong, Vo, Ba-Ngu
Format: Conference Paper
Published: 2017
Online Access:http://hdl.handle.net/20.500.11937/67528
_version_ 1848761589343191040
author Beard, Michael
Vo, Ba Tuong
Vo, Ba-Ngu
author_facet Beard, Michael
Vo, Ba Tuong
Vo, Ba-Ngu
author_sort Beard, Michael
building Curtin Institutional Repository
collection Online Access
description © 2017 IEEE. The optimal sub-pattern assignment (OSPA) metric is a distance between two sets of points that jointly accounts for the dissimilarity in the number of points and the values of the points in the respective sets. The OSPA metric is often used for measuring the distance between two sets of points in Euclidean space. A common example is in multi-target filtering, where the aim is to estimate the set of current target states, all of which have the same dimension. In multi-target tracking (MTT), the aim is to estimate the set of target tracks over a period of time, rather than the set of target states at each time step. In this case, it is not sufficient to analyse the multi-target filtering error at each time step in isolation. It is important that a metric for evaluating MTT performance accounts for the dissimilarity between the overall target tracks, which are generally of different dimensions. In this paper, we demonstrate that MTT error can be captured using the OSPA metric to define a distance between two sets of tracks.
first_indexed 2025-11-14T10:34:04Z
format Conference Paper
id curtin-20.500.11937-67528
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T10:34:04Z
publishDate 2017
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-675282023-08-02T06:39:11Z OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance Beard, Michael Vo, Ba Tuong Vo, Ba-Ngu © 2017 IEEE. The optimal sub-pattern assignment (OSPA) metric is a distance between two sets of points that jointly accounts for the dissimilarity in the number of points and the values of the points in the respective sets. The OSPA metric is often used for measuring the distance between two sets of points in Euclidean space. A common example is in multi-target filtering, where the aim is to estimate the set of current target states, all of which have the same dimension. In multi-target tracking (MTT), the aim is to estimate the set of target tracks over a period of time, rather than the set of target states at each time step. In this case, it is not sufficient to analyse the multi-target filtering error at each time step in isolation. It is important that a metric for evaluating MTT performance accounts for the dissimilarity between the overall target tracks, which are generally of different dimensions. In this paper, we demonstrate that MTT error can be captured using the OSPA metric to define a distance between two sets of tracks. 2017 Conference Paper http://hdl.handle.net/20.500.11937/67528 10.1109/ICCAIS.2017.8217598 restricted
spellingShingle Beard, Michael
Vo, Ba Tuong
Vo, Ba-Ngu
OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance
title OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance
title_full OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance
title_fullStr OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance
title_full_unstemmed OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance
title_short OSPA(2): Using the OSPA metric to evaluate multi-target tracking performance
title_sort ospa(2): using the ospa metric to evaluate multi-target tracking performance
url http://hdl.handle.net/20.500.11937/67528