Abstract

We propose a new framework that extends the standard Probability Hypothesis Density (PHD) filter for multiple targets having N⩾2 different types based on Random Finite Set theory, taking into account not only background clutter, but also confusions among detections of different target types, which are in general different in character from background clutter. Under Gaussianity and linearity assumptions, our framework extends the existing Gaussian mixture (GM) implementation of the standard PHD filter to create a N-type GM-PHD filter. The methodology is applied to real video sequences by integrating object detectors’ information into this filter for two scenarios. For both cases, Munkres's variant of the Hungarian assignment algorithm is used to associate tracked target identities between frames. This approach is evaluated and compared to both raw detection and independent GM-PHD filters using the Optimal Sub-pattern Assignment metric and discrimination rate. This shows the improved performance of our strategy on real video sequences.
LanguageEnglish
Pages257-271
Number of pages15
JournalJournal of Visual Communication and Image Representation
Volume59
Early online date17 Jan 2019
DOIs
StatePublished - Feb 2019

Fingerprint

Set theory
Detectors

Cite this

@article{2e9177383ed34e27af739657f8645048,
title = "Development of a N-type GM-PHD Filter for Multiple Target, Multiple Type Visual Tracking",
abstract = "We propose a new framework that extends the standard Probability Hypothesis Density (PHD) filter for multiple targets having N⩾2 different types based on Random Finite Set theory, taking into account not only background clutter, but also confusions among detections of different target types, which are in general different in character from background clutter. Under Gaussianity and linearity assumptions, our framework extends the existing Gaussian mixture (GM) implementation of the standard PHD filter to create a N-type GM-PHD filter. The methodology is applied to real video sequences by integrating object detectors’ information into this filter for two scenarios. For both cases, Munkres's variant of the Hungarian assignment algorithm is used to associate tracked target identities between frames. This approach is evaluated and compared to both raw detection and independent GM-PHD filters using the Optimal Sub-pattern Assignment metric and discrimination rate. This shows the improved performance of our strategy on real video sequences.",
author = "Baisa, {Nathanael L.} and Andrew Wallace",
year = "2019",
month = "2",
doi = "10.1016/j.jvcir.2019.01.026",
language = "English",
volume = "59",
pages = "257--271",
journal = "Journal of Visual Communication and Image Representation",
issn = "1047-3203",
publisher = "Academic Press Inc.",

}

Development of a N-type GM-PHD Filter for Multiple Target, Multiple Type Visual Tracking. / Baisa, Nathanael L.; Wallace, Andrew.

In: Journal of Visual Communication and Image Representation, Vol. 59, 02.2019, p. 257-271.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development of a N-type GM-PHD Filter for Multiple Target, Multiple Type Visual Tracking

AU - Baisa,Nathanael L.

AU - Wallace,Andrew

PY - 2019/2

Y1 - 2019/2

N2 - We propose a new framework that extends the standard Probability Hypothesis Density (PHD) filter for multiple targets having N⩾2 different types based on Random Finite Set theory, taking into account not only background clutter, but also confusions among detections of different target types, which are in general different in character from background clutter. Under Gaussianity and linearity assumptions, our framework extends the existing Gaussian mixture (GM) implementation of the standard PHD filter to create a N-type GM-PHD filter. The methodology is applied to real video sequences by integrating object detectors’ information into this filter for two scenarios. For both cases, Munkres's variant of the Hungarian assignment algorithm is used to associate tracked target identities between frames. This approach is evaluated and compared to both raw detection and independent GM-PHD filters using the Optimal Sub-pattern Assignment metric and discrimination rate. This shows the improved performance of our strategy on real video sequences.

AB - We propose a new framework that extends the standard Probability Hypothesis Density (PHD) filter for multiple targets having N⩾2 different types based on Random Finite Set theory, taking into account not only background clutter, but also confusions among detections of different target types, which are in general different in character from background clutter. Under Gaussianity and linearity assumptions, our framework extends the existing Gaussian mixture (GM) implementation of the standard PHD filter to create a N-type GM-PHD filter. The methodology is applied to real video sequences by integrating object detectors’ information into this filter for two scenarios. For both cases, Munkres's variant of the Hungarian assignment algorithm is used to associate tracked target identities between frames. This approach is evaluated and compared to both raw detection and independent GM-PHD filters using the Optimal Sub-pattern Assignment metric and discrimination rate. This shows the improved performance of our strategy on real video sequences.

U2 - 10.1016/j.jvcir.2019.01.026

DO - 10.1016/j.jvcir.2019.01.026

M3 - Article

VL - 59

SP - 257

EP - 271

JO - Journal of Visual Communication and Image Representation

T2 - Journal of Visual Communication and Image Representation

JF - Journal of Visual Communication and Image Representation

SN - 1047-3203

ER -