Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance

Xinyu Xu, Baoxin Li

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

Abstract

In the visual tracking domain, Particle Filtering (PF) can become quite inefficient when being applied into high dimensional state space. Rao-Blackwellisation [1] has been shown to be an effective method to reduce the size of the state space by marginalizing out some of the variables analytically [2]. In this paper based on our previous work [3] we proposed RBPF tracking algorithm with adaptive system noise model. Experiments using both simulation data and real data show that the proposed RBPF algorithm with adaptive noise variance improves its performance significantly over conventional Particle Filter tracking algorithm. The improvements manifest in three aspects: increased estimation accuracy, reduced variance for estimates and reduced particle numbers are needed to achieve the same level of accuracy.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
Volume6077
DOIs
StatePublished - 2006
EventVisual Communications and Image Processing 2006 - San Jose, CA, United States
Duration: Jan 17 2006Jan 19 2006

Other

OtherVisual Communications and Image Processing 2006
CountryUnited States
CitySan Jose, CA
Period1/17/061/19/06

Fingerprint

Adaptive systems
surveillance
filters
evaluation
tracking filters
optical tracking
data simulation
estimates
Experiments

Keywords

  • Adaptive noise
  • Rao-Blackwellised particle filter
  • Surveillance
  • Tracking

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Xu, X., & Li, B. (2006). Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 6077). [60770W] https://doi.org/10.1117/12.643073

Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance. / Xu, Xinyu; Li, Baoxin.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6077 2006. 60770W.

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

Xu, X & Li, B 2006, Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 6077, 60770W, Visual Communications and Image Processing 2006, San Jose, CA, United States, 1/17/06. https://doi.org/10.1117/12.643073
Xu X, Li B. Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6077. 2006. 60770W https://doi.org/10.1117/12.643073
Xu, Xinyu ; Li, Baoxin. / Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 6077 2006.
@inproceedings{be7cc4176cd54539afb4726a0fd1fd21,
title = "Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance",
abstract = "In the visual tracking domain, Particle Filtering (PF) can become quite inefficient when being applied into high dimensional state space. Rao-Blackwellisation [1] has been shown to be an effective method to reduce the size of the state space by marginalizing out some of the variables analytically [2]. In this paper based on our previous work [3] we proposed RBPF tracking algorithm with adaptive system noise model. Experiments using both simulation data and real data show that the proposed RBPF algorithm with adaptive noise variance improves its performance significantly over conventional Particle Filter tracking algorithm. The improvements manifest in three aspects: increased estimation accuracy, reduced variance for estimates and reduced particle numbers are needed to achieve the same level of accuracy.",
keywords = "Adaptive noise, Rao-Blackwellised particle filter, Surveillance, Tracking",
author = "Xinyu Xu and Baoxin Li",
year = "2006",
doi = "10.1117/12.643073",
language = "English (US)",
isbn = "0819461172",
volume = "6077",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

}

TY - GEN

T1 - Rao-Blackwellised particle filter with adaptive system noise and its evaluation for tracking in surveillance

AU - Xu, Xinyu

AU - Li, Baoxin

PY - 2006

Y1 - 2006

N2 - In the visual tracking domain, Particle Filtering (PF) can become quite inefficient when being applied into high dimensional state space. Rao-Blackwellisation [1] has been shown to be an effective method to reduce the size of the state space by marginalizing out some of the variables analytically [2]. In this paper based on our previous work [3] we proposed RBPF tracking algorithm with adaptive system noise model. Experiments using both simulation data and real data show that the proposed RBPF algorithm with adaptive noise variance improves its performance significantly over conventional Particle Filter tracking algorithm. The improvements manifest in three aspects: increased estimation accuracy, reduced variance for estimates and reduced particle numbers are needed to achieve the same level of accuracy.

AB - In the visual tracking domain, Particle Filtering (PF) can become quite inefficient when being applied into high dimensional state space. Rao-Blackwellisation [1] has been shown to be an effective method to reduce the size of the state space by marginalizing out some of the variables analytically [2]. In this paper based on our previous work [3] we proposed RBPF tracking algorithm with adaptive system noise model. Experiments using both simulation data and real data show that the proposed RBPF algorithm with adaptive noise variance improves its performance significantly over conventional Particle Filter tracking algorithm. The improvements manifest in three aspects: increased estimation accuracy, reduced variance for estimates and reduced particle numbers are needed to achieve the same level of accuracy.

KW - Adaptive noise

KW - Rao-Blackwellised particle filter

KW - Surveillance

KW - Tracking

UR - http://www.scopus.com/inward/record.url?scp=33645976463&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645976463&partnerID=8YFLogxK

U2 - 10.1117/12.643073

DO - 10.1117/12.643073

M3 - Conference contribution

SN - 0819461172

SN - 9780819461179

SN - 0819461172

SN - 9780819461179

VL - 6077

BT - Proceedings of SPIE - The International Society for Optical Engineering

ER -