Appearance tracking for video surveillance

Javier Varona; Jordi Gonzàlez; F. Xavier Roca; J. J. Villanueva

Appearance tracking for video surveillance

Javier Varona, Jordi Gonzàlez, F. Xavier Roca, J. J. Villanueva

Research output: Contribution to journal › Article › Research › peer-review

Abstract

We present an algorithm which tracks multiple objects for video surveillance applications. This algorithm is based on a Bayesian framework and a Particle filter. In order to use this method in practical applications we define a statistical model of the object appearance to build a robust likelihood function. The tracking process is only based on image data, therefore, a previous step to learn the object shape and their motion parameters is not necessary. Using the localization results, we can define a prior density which is used to initialize the algorithm. Finally, our method has been proved successfully in several sequences and its performance is more accurate than classical filters. © Springer-Verlag Berlin Heidelberg 2003.

Original language	English
Pages (from-to)	1041-1048
Journal	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	2652
Publication status	Published - 1 Dec 2003

Fingerprint Dive into the research topics of 'Appearance tracking for video surveillance'. Together they form a unique fingerprint.

Cite this

@article{bb4f44b40fd84079bfb496340c156f14,

title = "Appearance tracking for video surveillance",

abstract = "We present an algorithm which tracks multiple objects for video surveillance applications. This algorithm is based on a Bayesian framework and a Particle filter. In order to use this method in practical applications we define a statistical model of the object appearance to build a robust likelihood function. The tracking process is only based on image data, therefore, a previous step to learn the object shape and their motion parameters is not necessary. Using the localization results, we can define a prior density which is used to initialize the algorithm. Finally, our method has been proved successfully in several sequences and its performance is more accurate than classical filters. {\textcopyright} Springer-Verlag Berlin Heidelberg 2003.",

author = "Javier Varona and Jordi Gonz{\`a}lez and Roca, {F. Xavier} and Villanueva, {J. J.}",

year = "2003",

month = dec,

day = "1",

language = "English",

volume = "2652",

pages = "1041--1048",

}

Appearance tracking for video surveillance. / Varona, Javier; Gonzàlez, Jordi ; Roca, F. Xavier; Villanueva, J. J.

In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 2652, 01.12.2003, p. 1041-1048.

Research output: Contribution to journal › Article › Research › peer-review

TY - JOUR

T1 - Appearance tracking for video surveillance

AU - Varona, Javier

AU - Gonzàlez, Jordi

AU - Roca, F. Xavier

AU - Villanueva, J. J.

PY - 2003/12/1

Y1 - 2003/12/1

N2 - We present an algorithm which tracks multiple objects for video surveillance applications. This algorithm is based on a Bayesian framework and a Particle filter. In order to use this method in practical applications we define a statistical model of the object appearance to build a robust likelihood function. The tracking process is only based on image data, therefore, a previous step to learn the object shape and their motion parameters is not necessary. Using the localization results, we can define a prior density which is used to initialize the algorithm. Finally, our method has been proved successfully in several sequences and its performance is more accurate than classical filters. © Springer-Verlag Berlin Heidelberg 2003.

AB - We present an algorithm which tracks multiple objects for video surveillance applications. This algorithm is based on a Bayesian framework and a Particle filter. In order to use this method in practical applications we define a statistical model of the object appearance to build a robust likelihood function. The tracking process is only based on image data, therefore, a previous step to learn the object shape and their motion parameters is not necessary. Using the localization results, we can define a prior density which is used to initialize the algorithm. Finally, our method has been proved successfully in several sequences and its performance is more accurate than classical filters. © Springer-Verlag Berlin Heidelberg 2003.

M3 - Article

VL - 2652

SP - 1041

EP - 1048

ER -