Improved proposal distribution with gradient measures for tracking

PA Brasnett; LS Mihaylova; DR Bull; CN Canagarajah

doi:10.1109/MLSP.2005.1532883

Improved proposal distribution with gradient measures for tracking

PA Brasnett, LS Mihaylova, DR Bull, CN Canagarajah

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

3 Citations (Scopus)

357 Downloads (Pure)

Abstract

Particle filters have become a useful tool for the task of object tracking due to their applicability to a wide range of situations. To be able to obtain an accurate estimate from a particle filter a large number of particles is usually necessary. A crucial step in the design of a particle filter is the choice of the proposal distribution. A common choice for the proposal distribution is to use the transition distribution which models the dynamics of the system but takes no account of the current measurements. We present a particle filter for tracking rigid objects in video sequences that makes use of image gradients in the current frame to improve the proposal distribution. The gradient information is efficiently incorporated in the filter to minimise the computational cost. Results from synthetic and natural sequences show that the gradient information improves the accuracy and reduces the number of particles required

Translated title of the contribution	Improved proposal distribution with gradient measures for tracking
Original language	English
Title of host publication	IEEE Workshop on Machine Learning for Signal Processing, Mystic, CT, United States
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	105 - 110
Number of pages	6
ISBN (Print)	0780395174
DOIs	https://doi.org/10.1109/MLSP.2005.1532883
Publication status	Published - Sept 2005
Event	IEEE Workshop on Machine Learning for Signal Processing - Mystic, Connecticut, United States Duration: 28 Sept 2005 → 30 Sept 2005 http://mlsp2005.conwiz.dk/

Conference

Conference	IEEE Workshop on Machine Learning for Signal Processing
Country/Territory	United States
City	Mystic, Connecticut
Period	28/09/05 → 30/09/05
Internet address	http://mlsp2005.conwiz.dk/

Bibliographical note

Rose publication type: Conference contribution

Sponsorship: This work has been conducted with support from the UK MOD Data and Information Fusion Defence Technology Centre under project DIF DTC 2.2.

Terms of use: Copyright © 2005 IEEE. Reprinted from IEEE Workshop on Machine Learning for Signal Processing, 2005.

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 [email protected].

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

Access to Document

10.1109/MLSP.2005.1532883

Brasnett IEEE machine 2005Accepted author manuscript, 280 KB

http://hdl.handle.net/1983/1507

Handle.net

Persistent link

Cite this

@inproceedings{17ed890c1351406eb003e42c46b3040e,

title = "Improved proposal distribution with gradient measures for tracking",

abstract = "Particle filters have become a useful tool for the task of object tracking due to their applicability to a wide range of situations. To be able to obtain an accurate estimate from a particle filter a large number of particles is usually necessary. A crucial step in the design of a particle filter is the choice of the proposal distribution. A common choice for the proposal distribution is to use the transition distribution which models the dynamics of the system but takes no account of the current measurements. We present a particle filter for tracking rigid objects in video sequences that makes use of image gradients in the current frame to improve the proposal distribution. The gradient information is efficiently incorporated in the filter to minimise the computational cost. Results from synthetic and natural sequences show that the gradient information improves the accuracy and reduces the number of particles required",

author = "PA Brasnett and LS Mihaylova and DR Bull and CN Canagarajah",

note = "Rose publication type: Conference contribution Sponsorship: This work has been conducted with support from the UK MOD Data and Information Fusion Defence Technology Centre under project DIF DTC 2.2. Terms of use: Copyright {\textcopyright} 2005 IEEE. Reprinted from IEEE Workshop on Machine Learning for Signal Processing, 2005. 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 [email protected]. By choosing to view this document, you agree to all provisions of the copyright laws protecting it. ; IEEE Workshop on Machine Learning for Signal Processing ; Conference date: 28-09-2005 Through 30-09-2005",

year = "2005",

month = sep,

doi = "10.1109/MLSP.2005.1532883",

language = "English",

isbn = "0780395174",

pages = "105 -- 110",

booktitle = "IEEE Workshop on Machine Learning for Signal Processing, Mystic, CT, United States",

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

address = "United States",

url = "http://mlsp2005.conwiz.dk/",

}

Brasnett, PA, Mihaylova, LS, Bull, DR & Canagarajah, CN 2005, Improved proposal distribution with gradient measures for tracking. in IEEE Workshop on Machine Learning for Signal Processing, Mystic, CT, United States. Institute of Electrical and Electronics Engineers (IEEE), pp. 105 - 110, IEEE Workshop on Machine Learning for Signal Processing, Mystic, Connecticut, United States, 28/09/05. https://doi.org/10.1109/MLSP.2005.1532883

Improved proposal distribution with gradient measures for tracking. / Brasnett, PA; Mihaylova, LS; Bull, DR et al.
IEEE Workshop on Machine Learning for Signal Processing, Mystic, CT, United States. Institute of Electrical and Electronics Engineers (IEEE), 2005. p. 105 - 110.

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

TY - GEN

T1 - Improved proposal distribution with gradient measures for tracking

AU - Brasnett, PA

AU - Mihaylova, LS

AU - Bull, DR

AU - Canagarajah, CN

N1 - Rose publication type: Conference contribution Sponsorship: This work has been conducted with support from the UK MOD Data and Information Fusion Defence Technology Centre under project DIF DTC 2.2. Terms of use: Copyright © 2005 IEEE. Reprinted from IEEE Workshop on Machine Learning for Signal Processing, 2005. 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 [email protected]. By choosing to view this document, you agree to all provisions of the copyright laws protecting it.

PY - 2005/9

Y1 - 2005/9

N2 - Particle filters have become a useful tool for the task of object tracking due to their applicability to a wide range of situations. To be able to obtain an accurate estimate from a particle filter a large number of particles is usually necessary. A crucial step in the design of a particle filter is the choice of the proposal distribution. A common choice for the proposal distribution is to use the transition distribution which models the dynamics of the system but takes no account of the current measurements. We present a particle filter for tracking rigid objects in video sequences that makes use of image gradients in the current frame to improve the proposal distribution. The gradient information is efficiently incorporated in the filter to minimise the computational cost. Results from synthetic and natural sequences show that the gradient information improves the accuracy and reduces the number of particles required

AB - Particle filters have become a useful tool for the task of object tracking due to their applicability to a wide range of situations. To be able to obtain an accurate estimate from a particle filter a large number of particles is usually necessary. A crucial step in the design of a particle filter is the choice of the proposal distribution. A common choice for the proposal distribution is to use the transition distribution which models the dynamics of the system but takes no account of the current measurements. We present a particle filter for tracking rigid objects in video sequences that makes use of image gradients in the current frame to improve the proposal distribution. The gradient information is efficiently incorporated in the filter to minimise the computational cost. Results from synthetic and natural sequences show that the gradient information improves the accuracy and reduces the number of particles required

U2 - 10.1109/MLSP.2005.1532883

DO - 10.1109/MLSP.2005.1532883

M3 - Conference Contribution (Conference Proceeding)

SN - 0780395174

SP - 105

EP - 110

BT - IEEE Workshop on Machine Learning for Signal Processing, Mystic, CT, United States

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - IEEE Workshop on Machine Learning for Signal Processing

Y2 - 28 September 2005 through 30 September 2005

ER -

Improved proposal distribution with gradient measures for tracking

Abstract

Conference

Bibliographical note

Access to Document

Handle.net

Fingerprint

Cite this