Multiple Target, Multiple Type Visual Tracking using a Tri-GM-PHD Filter

Nathanael Lemessa Baisa, Andrew Michael Wallace

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

We propose a new framework that extends the standard Probability Hypothesis Density (PHD) filter for multiple targets having three different types, taking into account not only background false positives (clutter), but also confusion etween detections of different target types, which are in general different in character from background clutter. Our framework extends the existing Gaussian Mixture (GM) implementation of the PHD filter to create a tri-GM-PHD filter based on Random Finite Set (RFS) theory. The methodology is applied to real video sequences containing three types of multiple targets in the same scene, two football teams and a referee, using separate detections. Subsequently, unkMres’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 detections and independent GM-PHD filters using the Optimal Sub-pattern Assignment (OSPA) metric and discrimination rate. This shows the improved performance of our strategy on real video sequences.
Original languageEnglish
Title of host publicationProceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
PublisherScience and Technology Publications, Lda
Pages467-477
Number of pages11
Volume6
ISBN (Print)9789897582271
DOIs
Publication statusPublished - 2017
Event12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Porto, Portugal
Duration: 27 Feb 20172 Mar 2017
http://www.visapp.visigrapp.org/?y=2017

Conference

Conference12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Abbreviated titleVISAPP2017
CountryPortugal
CityPorto
Period27/02/172/03/17
Internet address

Fingerprint Dive into the research topics of 'Multiple Target, Multiple Type Visual Tracking using a Tri-GM-PHD Filter'. Together they form a unique fingerprint.

  • Cite this

    Baisa, N. L., & Wallace, A. M. (2017). Multiple Target, Multiple Type Visual Tracking using a Tri-GM-PHD Filter. In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (Vol. 6, pp. 467-477). Science and Technology Publications, Lda . https://doi.org/10.5220/0006145704670477