A hierarchical dynamic bayesian network approach to visual tracking

Authors:
Hua Li;Rong Xiao;Hong-Jiang Zhang;Li-Zhong Peng
Affiliations:
LMAM, Department of Mathematics, School of Mathematical Sciences, Peking University, Beijing, P. R. China;Microsoft Research Asia, Beijing, P. R. China;Microsoft Research Asia, Beijing, P. R. China;LMAM, Department of Mathematics, School of Mathematical Sciences, Peking University, Beijing, P. R. China
Venue:
PCM'04 Proceedings of the 5th Pacific Rim Conference on Advances in Multimedia Information Processing - Volume Part II
Year:
2004

Citing 7
Cited 0

Finite-dimensional filters for passive tracking of Markov jump linear systems

Automatica (Journal of IFAC)
CONDENSATION—Conditional Density Propagation forVisual Tracking

International Journal of Computer Vision
On sequential Monte Carlo sampling methods for Bayesian filtering

Statistics and Computing
Contour Tracking by Stochastic Propagation of Conditional Density

ECCV '96 Proceedings of the 4th European Conference on Computer Vision-Volume I - Volume I
ICONDENSATION: Unifying Low-Level and High-Level Tracking in a Stochastic Framework

ECCV '98 Proceedings of the 5th European Conference on Computer Vision-Volume I - Volume I
Colour Model Selection and Adaption in Dynamic Scenes

ECCV '98 Proceedings of the 5th European Conference on Computer Vision-Volume I - Volume I
Switching observation models for contour tracking in clutter

CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

Usually a uniform observation strategy will result in frustrated tracking processes. To address this problem, we construct a flexible model with Hierarchical Dynamic Bayesian Network by introducing hidden variables to infer the intrinsic properties of the state and observation spaces. With this model, a dynamic-mapping is built between target state space and the observation space. Based on a decoupling based inference strategy, a tractable solution for this algorithm is proposed. Experiments of human face tracking under various poses and occlusions show promising results.