Components analysis of hidden Markov models in computer vision

Hidden Markov models (HMMs) have become a standard tool for pattern recognition in computer vision. Although parameter and topology estimation have been studied, and still are, detailed analysis of how these estimated parameters contribute to HMM performance is rarely addressed. We develop tools for measuring such contributions and illustrate key issues in a representative task of gesture recognition - 3D motion recovery from 2D projections.

[1]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[2]  D. Vere-Jones.  Markov Chains , 1972, Nature.

[3]  L. Goddard.  Information Theory , 1962, Nature.

[4]  Matthew Brand,et al.  Discovery and Segmentation of Activities in Video , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Terry Caelli,et al.  On Learning the Shape of Complex Actions , 2001, IWVF.

[6]  Terry Caelli,et al.  Shape Tracking and Production Using Hidden Markov Models , 2001, Int. J. Pattern Recognit. Artif. Intell..

[7]  Michael I. Jordan,et al.  Factorial Hidden Markov Models , 1995, Machine Learning.

[8]  Gene H. Golub,et al.  Matrix computations , 1983 .