Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 30174
Human Body Configuration using Bayesian Model

Authors: Rui. Zhang, Yiming. Pi

Abstract:

In this paper we present a novel approach for human Body configuration based on the Silhouette. We propose to address this problem under the Bayesian framework. We use an effective Model based MCMC (Markov Chain Monte Carlo) method to solve the configuration problem, in which the best configuration could be defined as MAP (maximize a posteriori probability) in Bayesian model. This model based MCMC utilizes the human body model to drive the MCMC sampling from the solution space. It converses the original high dimension space into a restricted sub-space constructed by the human model and uses a hybrid sampling algorithm. We choose an explicit human model and carefully select the likelihood functions to represent the best configuration solution. The experiments show that this method could get an accurate configuration and timesaving for different human from multi-views.

Keywords: Bayesian framework, MCMC, model based, human body configuration.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1333350

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 919

References:


[1] P.F.Felzenswalb and D.P.Huttenlocher, "Efficient matching of pictorial structure" in Proceeding of the computer vision and pattern Recognition, 2000, pp. 66-73.
[2] K.Rohr, "Towards model based recognition of human movements in image sequence" in CVGIP, 1994, vol.59
[3] Y.Song, X.Feng and P.perona, "Towards detection of human motion". in Proceeding of the computer vision and pattern Recognition, 2000
[4] D.M.Gavrila and L.Davis, "Towards 3-D model-based Tracking and recognition of Human Movement: a Multi-view Approach" In IEEE international conference on automatic Face and Gesture Recognition.1995
[5] S.Niyogi and E.Adelson, "Analyzing and recognizing walking figures in XYT", in Proceeding of the computer vision and pattern Recognition. 1994, pp. 469-474.
[6] Robert.T.Collins, Ralph Gross and Jianbo shi, "Silhouette-based Human Identification from Body Shape and Gait" In IEEE international conference on automatic Face and Gesture Recognition 2002
[7] R.Rosales and S.Sclaroff, "Inferring Body Pose without Tracking Body Parts". in Proceeding of the computer vision and pattern Recognition, 2000.
[8] C.Andrieu, N.De.Freitas,A.Doucet and M.jordan, "An Introduction to MCMC for Machine Learning". Machine Learning, 50, 5-43, 2003
[9] W.K.Hasting, "Monte Carlo Sampling Methods Using Markov Chain and Their Application". Biometrika, 1970.
[10] T. Zhao and R. Nevatia. "Bayesian Human Segmentation in Crowded Situations", in Proceeding of the computer vision and pattern Recognition, pp.459-466. 2003
[11] W.Gilk, S.Richardson and D.Spiegelhalter, Markov Chain Monte Carlo in Practice Chapman and Hall, 1996
[12] B.Walsh, Markov Chain Monte Carlo and Gibbs sampling. Lecture Notes for EEB 596z, 2002.
[13] M. Hason "Tutorial on Markov Chain Monte Carlo" Technical report, Los Alamos National laboratory, 2000.