Object Tracking Using Point Matching Based on MCMC

Author

Hongbo, Yang ; Xia, Hou

Author_Institution

Sch. of Autom., Beijing Inf. Sci. & Technol. Univ., Beijing, China

Volume

4

fYear

2009

fDate

March 31 2009-April 2 2009

Firstpage

182

Lastpage

186

Abstract

Most of the reported object tracking methods achieved by estimating the object position cannot suit for large image deformations. For this reason, a new object tracking base on MCMC point matching method is presented in this paper. This method applies MCMC algorithm to solve the posterior probability distribution problem and obtain the optimal matching parameters include position, rotation, scale information. The test results prove this method can successfully cope with target scale or orientation variations in video tracking.

Keywords

Markov processes; Monte Carlo methods; image matching; image sampling; object detection; statistical distributions; tracking; video signal processing; MCMC point matching method; MCMC sampling method; Markov Chain Monte Carlo algorithm; object video tracking method; optimal matching parameter; posterior probability distribution problem; Bayesian methods; Computer science; Image motion analysis; Information science; Matrix decomposition; Optical distortion; Optical filters; Pixel; Probability distribution; Target tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science and Information Engineering, 2009 WRI World Congress on

Conference_Location

Los Angeles, CA

Print_ISBN

978-0-7695-3507-4

Type

conf

DOI

10.1109/CSIE.2009.319

Filename

5170984