Monte Carlo simulation techniques for probabilistic tracking

Baoxin Li, Rama Chellappa, Hankyu Moon

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

Abstract

In this paper, two novel approaches to probabilistic tracking using Monte Carlo simulation are presented. The first approach is a 3D shape encoded object tracking algorithm. The measurements are derived using the outputs of shape-encoded filters. The nonlinear state estimation is performed by solving the Zakai equation, and we use the branching particle propagation method for computing the solution. The second approach is an algorithm for simultaneous tracking and verification in video data. The approach is based on posterior density estimation using sequential Monte Carlo methods. Visual tracking, which is in essence a temporal correspondence problem, is solved through probability density propagation, with the density being defined over a proper state space characterizing the object configuration. Verification is realized through hypothesis testing using the estimated posterior density. Several applications of both approaches including human head and body tracking, human identification and facial feature based face verification are illustrated by experiments devised to evaluate their performance.

Original languageEnglish (US)
Title of host publicationConference Record of the Asilomar Conference on Signals, Systems and Computers
EditorsM.B. Matthews
Pages75-82
Number of pages8
Volume1
StatePublished - 2001
Externally publishedYes
Event35th Asilomar Conference on Signals, Systems and Computers - Pacific Grove, CA, United States
Duration: Nov 4 2001Nov 7 2001

Other

Other35th Asilomar Conference on Signals, Systems and Computers
CountryUnited States
CityPacific Grove, CA
Period11/4/0111/7/01

Fingerprint

State estimation
Monte Carlo methods
Testing
Monte Carlo simulation
Experiments

ASJC Scopus subject areas

  • Hardware and Architecture
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Li, B., Chellappa, R., & Moon, H. (2001). Monte Carlo simulation techniques for probabilistic tracking. In M. B. Matthews (Ed.), Conference Record of the Asilomar Conference on Signals, Systems and Computers (Vol. 1, pp. 75-82)

Monte Carlo simulation techniques for probabilistic tracking. / Li, Baoxin; Chellappa, Rama; Moon, Hankyu.

Conference Record of the Asilomar Conference on Signals, Systems and Computers. ed. / M.B. Matthews. Vol. 1 2001. p. 75-82.

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

Li, B, Chellappa, R & Moon, H 2001, Monte Carlo simulation techniques for probabilistic tracking. in MB Matthews (ed.), Conference Record of the Asilomar Conference on Signals, Systems and Computers. vol. 1, pp. 75-82, 35th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, United States, 11/4/01.
Li B, Chellappa R, Moon H. Monte Carlo simulation techniques for probabilistic tracking. In Matthews MB, editor, Conference Record of the Asilomar Conference on Signals, Systems and Computers. Vol. 1. 2001. p. 75-82
Li, Baoxin ; Chellappa, Rama ; Moon, Hankyu. / Monte Carlo simulation techniques for probabilistic tracking. Conference Record of the Asilomar Conference on Signals, Systems and Computers. editor / M.B. Matthews. Vol. 1 2001. pp. 75-82
@inproceedings{df747da8ab8548b2aae1d8348ad5d2fe,
title = "Monte Carlo simulation techniques for probabilistic tracking",
abstract = "In this paper, two novel approaches to probabilistic tracking using Monte Carlo simulation are presented. The first approach is a 3D shape encoded object tracking algorithm. The measurements are derived using the outputs of shape-encoded filters. The nonlinear state estimation is performed by solving the Zakai equation, and we use the branching particle propagation method for computing the solution. The second approach is an algorithm for simultaneous tracking and verification in video data. The approach is based on posterior density estimation using sequential Monte Carlo methods. Visual tracking, which is in essence a temporal correspondence problem, is solved through probability density propagation, with the density being defined over a proper state space characterizing the object configuration. Verification is realized through hypothesis testing using the estimated posterior density. Several applications of both approaches including human head and body tracking, human identification and facial feature based face verification are illustrated by experiments devised to evaluate their performance.",
author = "Baoxin Li and Rama Chellappa and Hankyu Moon",
year = "2001",
language = "English (US)",
volume = "1",
pages = "75--82",
editor = "M.B. Matthews",
booktitle = "Conference Record of the Asilomar Conference on Signals, Systems and Computers",

}

TY - GEN

T1 - Monte Carlo simulation techniques for probabilistic tracking

AU - Li, Baoxin

AU - Chellappa, Rama

AU - Moon, Hankyu

PY - 2001

Y1 - 2001

N2 - In this paper, two novel approaches to probabilistic tracking using Monte Carlo simulation are presented. The first approach is a 3D shape encoded object tracking algorithm. The measurements are derived using the outputs of shape-encoded filters. The nonlinear state estimation is performed by solving the Zakai equation, and we use the branching particle propagation method for computing the solution. The second approach is an algorithm for simultaneous tracking and verification in video data. The approach is based on posterior density estimation using sequential Monte Carlo methods. Visual tracking, which is in essence a temporal correspondence problem, is solved through probability density propagation, with the density being defined over a proper state space characterizing the object configuration. Verification is realized through hypothesis testing using the estimated posterior density. Several applications of both approaches including human head and body tracking, human identification and facial feature based face verification are illustrated by experiments devised to evaluate their performance.

AB - In this paper, two novel approaches to probabilistic tracking using Monte Carlo simulation are presented. The first approach is a 3D shape encoded object tracking algorithm. The measurements are derived using the outputs of shape-encoded filters. The nonlinear state estimation is performed by solving the Zakai equation, and we use the branching particle propagation method for computing the solution. The second approach is an algorithm for simultaneous tracking and verification in video data. The approach is based on posterior density estimation using sequential Monte Carlo methods. Visual tracking, which is in essence a temporal correspondence problem, is solved through probability density propagation, with the density being defined over a proper state space characterizing the object configuration. Verification is realized through hypothesis testing using the estimated posterior density. Several applications of both approaches including human head and body tracking, human identification and facial feature based face verification are illustrated by experiments devised to evaluate their performance.

UR - http://www.scopus.com/inward/record.url?scp=0035574194&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035574194&partnerID=8YFLogxK

M3 - Conference contribution

VL - 1

SP - 75

EP - 82

BT - Conference Record of the Asilomar Conference on Signals, Systems and Computers

A2 - Matthews, M.B.

ER -