A generic approach to object matching and tracking

Xiaokun Li, Chiman Kwan, Gang Mei, Baoxin Li

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

12 Citations (Scopus)

Abstract

In this paper, a generic approach to object matching and fast tracking in video and image sequence is presented. The approach first uses Gabor filters to extract flexible and reliable features as the basis of object matching and tracking. Then, a modified Elastic Graph Matching method is proposed for accurate object matching. A novel method based on posterior probability density estimation through sequential Monte Carlo method, called as Sequential Importance Sampling (SIS) method, is also developed to track multiple objects simultaneously. Several applications of our proposed approach are given for performance evaluation, which includes moving target tracking, stereo (3D) imaging, and camera stabilization. The experimental results demonstrated the efficacy of the approach which can also be applied to many other military and civilian applications, such as moving target verification and tracking, visual surveillance of public transportation, country border control, battlefield inspection and analysis, etc.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages839-849
Number of pages11
Volume4141 LNCS
StatePublished - 2006
Event3rd International Conference on Image Analysis and Recognition, ICIAR 2006 - Povoa de Varzim, Portugal
Duration: Sep 18 2006Sep 20 2006

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4141 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other3rd International Conference on Image Analysis and Recognition, ICIAR 2006
CountryPortugal
CityPovoa de Varzim
Period9/18/069/20/06

Fingerprint

Gabor filters
Importance sampling
Moving Target
Target tracking
Monte Carlo Method
Monte Carlo methods
Stabilization
Inspection
Cameras
Sequential Importance Sampling
Visual Surveillance
Sequential Monte Carlo Methods
Imaging techniques
Gabor Filter
Graph Matching
3D Imaging
Target Tracking
Posterior Probability
Density Estimation
Sampling Methods

Keywords

  • Feature extraction
  • Image analysis
  • Object matching
  • Real-time tracking

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this

Li, X., Kwan, C., Mei, G., & Li, B. (2006). A generic approach to object matching and tracking. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4141 LNCS, pp. 839-849). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4141 LNCS).

A generic approach to object matching and tracking. / Li, Xiaokun; Kwan, Chiman; Mei, Gang; Li, Baoxin.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 4141 LNCS 2006. p. 839-849 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4141 LNCS).

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

Li, X, Kwan, C, Mei, G & Li, B 2006, A generic approach to object matching and tracking. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 4141 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4141 LNCS, pp. 839-849, 3rd International Conference on Image Analysis and Recognition, ICIAR 2006, Povoa de Varzim, Portugal, 9/18/06.
Li X, Kwan C, Mei G, Li B. A generic approach to object matching and tracking. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 4141 LNCS. 2006. p. 839-849. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Li, Xiaokun ; Kwan, Chiman ; Mei, Gang ; Li, Baoxin. / A generic approach to object matching and tracking. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 4141 LNCS 2006. pp. 839-849 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{10c5e5c685b34fd69e71606090e4b2ed,
title = "A generic approach to object matching and tracking",
abstract = "In this paper, a generic approach to object matching and fast tracking in video and image sequence is presented. The approach first uses Gabor filters to extract flexible and reliable features as the basis of object matching and tracking. Then, a modified Elastic Graph Matching method is proposed for accurate object matching. A novel method based on posterior probability density estimation through sequential Monte Carlo method, called as Sequential Importance Sampling (SIS) method, is also developed to track multiple objects simultaneously. Several applications of our proposed approach are given for performance evaluation, which includes moving target tracking, stereo (3D) imaging, and camera stabilization. The experimental results demonstrated the efficacy of the approach which can also be applied to many other military and civilian applications, such as moving target verification and tracking, visual surveillance of public transportation, country border control, battlefield inspection and analysis, etc.",
keywords = "Feature extraction, Image analysis, Object matching, Real-time tracking",
author = "Xiaokun Li and Chiman Kwan and Gang Mei and Baoxin Li",
year = "2006",
language = "English (US)",
isbn = "3540448918",
volume = "4141 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "839--849",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - A generic approach to object matching and tracking

AU - Li, Xiaokun

AU - Kwan, Chiman

AU - Mei, Gang

AU - Li, Baoxin

PY - 2006

Y1 - 2006

N2 - In this paper, a generic approach to object matching and fast tracking in video and image sequence is presented. The approach first uses Gabor filters to extract flexible and reliable features as the basis of object matching and tracking. Then, a modified Elastic Graph Matching method is proposed for accurate object matching. A novel method based on posterior probability density estimation through sequential Monte Carlo method, called as Sequential Importance Sampling (SIS) method, is also developed to track multiple objects simultaneously. Several applications of our proposed approach are given for performance evaluation, which includes moving target tracking, stereo (3D) imaging, and camera stabilization. The experimental results demonstrated the efficacy of the approach which can also be applied to many other military and civilian applications, such as moving target verification and tracking, visual surveillance of public transportation, country border control, battlefield inspection and analysis, etc.

AB - In this paper, a generic approach to object matching and fast tracking in video and image sequence is presented. The approach first uses Gabor filters to extract flexible and reliable features as the basis of object matching and tracking. Then, a modified Elastic Graph Matching method is proposed for accurate object matching. A novel method based on posterior probability density estimation through sequential Monte Carlo method, called as Sequential Importance Sampling (SIS) method, is also developed to track multiple objects simultaneously. Several applications of our proposed approach are given for performance evaluation, which includes moving target tracking, stereo (3D) imaging, and camera stabilization. The experimental results demonstrated the efficacy of the approach which can also be applied to many other military and civilian applications, such as moving target verification and tracking, visual surveillance of public transportation, country border control, battlefield inspection and analysis, etc.

KW - Feature extraction

KW - Image analysis

KW - Object matching

KW - Real-time tracking

UR - http://www.scopus.com/inward/record.url?scp=33749675697&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749675697&partnerID=8YFLogxK

M3 - Conference contribution

SN - 3540448918

SN - 9783540448914

VL - 4141 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 839

EP - 849

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -