Multiple object tracking using particle filters

MH Jaward; LS Mihaylova; CN Canagarajah; DR Bull

doi:10.1109/AERO.2006.1655926

Multiple object tracking using particle filters

MH Jaward, LS Mihaylova, CN Canagarajah, DR Bull

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

33 Citations (Scopus)

358 Downloads (Pure)

Abstract

The particle filtering technique with multiple cues such as colour, texture and edges as observation features is a powerful technique for tracking deformable objects in image sequences with complex backgrounds. In this paper, our recent work (Brasnett et al., 2005) on single object tracking using particle filters is extended to multiple objects. In the proposed scheme, track initialisation is embedded in the particle filter without relying on an external object detection scheme. The proposed scheme avoids the use of hybrid state estimation for the estimation of number of active objects and its associated state vectors as proposed in (Czyz et al., 2005). The number of active objects and track management are handled by means of probabilities of the number of active objects in a given frame. These probabilities are shown to be easily estimated by the Monte Carlo data association algorithm used in our algorithm. The proposed particle filter (PF) embeds a data association technique based on the joint probabilistic data association (JPDA) which handles the uncertainty of the measurement origin. The algorithm is able to cope with partial occlusions and to recover the tracks after temporary loss. The probabilities calculated for data associations take part in the calculation of probabilities of the number of objects. We evaluate the performance of the proposed filter on various real-world video sequences with appearing and disappearing targets.

Translated title of the contribution	Multiple object tracking using particle filters
Original language	English
Title of host publication	2006 IEEE Aerospace Conference, Big Sky, MT, United States
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
ISBN (Print)	078039545X
DOIs	https://doi.org/10.1109/AERO.2006.1655926
Publication status	Published - Mar 2006
Event	IEEE Aerospace Conference - Big Sky, MT, United States Duration: 1 Mar 2006 → …

Conference

Conference	IEEE Aerospace Conference
Country	United States
City	Big Sky, MT
Period	1/03/06 → …

Bibliographical note

Rose publication type: Conference contribution

Sponsorship: The authors are grateful for the financial support from the UK MOD Data and Information Fusion Defence Technology Centre.

Terms of use: Copyright © 2006 IEEE. Reprinted from IEEE Aerospace Conference, 2006.

This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of the University of Bristol's products or services. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@ieee.org.

By choosing to view this document, you agree to all provisions of the copyright laws protecting it.

Access to Document

10.1109/AERO.2006.1655926

Jaward IEEE Aerospace 2006Accepted author manuscript, 9.34 MB

http://hdl.handle.net/1983/1427

Fingerprint Dive into the research topics of 'Multiple object tracking using particle filters'. Together they form a unique fingerprint.

Cite this

@inproceedings{7064720de7cc43b7bbba17b7f73dce5c,

title = "Multiple object tracking using particle filters",

abstract = "The particle filtering technique with multiple cues such as colour, texture and edges as observation features is a powerful technique for tracking deformable objects in image sequences with complex backgrounds. In this paper, our recent work (Brasnett et al., 2005) on single object tracking using particle filters is extended to multiple objects. In the proposed scheme, track initialisation is embedded in the particle filter without relying on an external object detection scheme. The proposed scheme avoids the use of hybrid state estimation for the estimation of number of active objects and its associated state vectors as proposed in (Czyz et al., 2005). The number of active objects and track management are handled by means of probabilities of the number of active objects in a given frame. These probabilities are shown to be easily estimated by the Monte Carlo data association algorithm used in our algorithm. The proposed particle filter (PF) embeds a data association technique based on the joint probabilistic data association (JPDA) which handles the uncertainty of the measurement origin. The algorithm is able to cope with partial occlusions and to recover the tracks after temporary loss. The probabilities calculated for data associations take part in the calculation of probabilities of the number of objects. We evaluate the performance of the proposed filter on various real-world video sequences with appearing and disappearing targets.",

author = "MH Jaward and LS Mihaylova and CN Canagarajah and DR Bull",

note = "Rose publication type: Conference contribution Sponsorship: The authors are grateful for the financial support from the UK MOD Data and Information Fusion Defence Technology Centre. Terms of use: Copyright {\textcopyright} 2006 IEEE. Reprinted from IEEE Aerospace Conference, 2006. This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of the University of Bristol's products or services. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@ieee.org. By choosing to view this document, you agree to all provisions of the copyright laws protecting it. ; IEEE Aerospace Conference ; Conference date: 01-03-2006",

year = "2006",

month = mar,

doi = "10.1109/AERO.2006.1655926",

language = "English",

isbn = "078039545X",

booktitle = "2006 IEEE Aerospace Conference, Big Sky, MT, United States",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

}

TY - GEN

T1 - Multiple object tracking using particle filters

AU - Jaward, MH

AU - Mihaylova, LS

AU - Canagarajah, CN

AU - Bull, DR

N1 - Rose publication type: Conference contribution Sponsorship: The authors are grateful for the financial support from the UK MOD Data and Information Fusion Defence Technology Centre. Terms of use: Copyright © 2006 IEEE. Reprinted from IEEE Aerospace Conference, 2006. This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of the University of Bristol's products or services. Internal or personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@ieee.org. By choosing to view this document, you agree to all provisions of the copyright laws protecting it.

PY - 2006/3

Y1 - 2006/3

N2 - The particle filtering technique with multiple cues such as colour, texture and edges as observation features is a powerful technique for tracking deformable objects in image sequences with complex backgrounds. In this paper, our recent work (Brasnett et al., 2005) on single object tracking using particle filters is extended to multiple objects. In the proposed scheme, track initialisation is embedded in the particle filter without relying on an external object detection scheme. The proposed scheme avoids the use of hybrid state estimation for the estimation of number of active objects and its associated state vectors as proposed in (Czyz et al., 2005). The number of active objects and track management are handled by means of probabilities of the number of active objects in a given frame. These probabilities are shown to be easily estimated by the Monte Carlo data association algorithm used in our algorithm. The proposed particle filter (PF) embeds a data association technique based on the joint probabilistic data association (JPDA) which handles the uncertainty of the measurement origin. The algorithm is able to cope with partial occlusions and to recover the tracks after temporary loss. The probabilities calculated for data associations take part in the calculation of probabilities of the number of objects. We evaluate the performance of the proposed filter on various real-world video sequences with appearing and disappearing targets.

AB - The particle filtering technique with multiple cues such as colour, texture and edges as observation features is a powerful technique for tracking deformable objects in image sequences with complex backgrounds. In this paper, our recent work (Brasnett et al., 2005) on single object tracking using particle filters is extended to multiple objects. In the proposed scheme, track initialisation is embedded in the particle filter without relying on an external object detection scheme. The proposed scheme avoids the use of hybrid state estimation for the estimation of number of active objects and its associated state vectors as proposed in (Czyz et al., 2005). The number of active objects and track management are handled by means of probabilities of the number of active objects in a given frame. These probabilities are shown to be easily estimated by the Monte Carlo data association algorithm used in our algorithm. The proposed particle filter (PF) embeds a data association technique based on the joint probabilistic data association (JPDA) which handles the uncertainty of the measurement origin. The algorithm is able to cope with partial occlusions and to recover the tracks after temporary loss. The probabilities calculated for data associations take part in the calculation of probabilities of the number of objects. We evaluate the performance of the proposed filter on various real-world video sequences with appearing and disappearing targets.

U2 - 10.1109/AERO.2006.1655926

DO - 10.1109/AERO.2006.1655926

M3 - Conference Contribution (Conference Proceeding)

SN - 078039545X

BT - 2006 IEEE Aerospace Conference, Big Sky, MT, United States

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - IEEE Aerospace Conference

Y2 - 1 March 2006

ER -