Title :
On merging hidden Markov models with deformable templates
Author :
Rao, Ram R. ; Mersereau, Russell M.
Author_Institution :
Sch. of Electr. & Comput. Eng., Georgia Inst. of Technol., Atlanta, GA, USA
Abstract :
Hidden Markov modeling has proven extremely useful for statistical analysis of speech signals. There are, however, inherent problems in two dimensional extensions to HMMs, one of which is the exponential complexity associated with fully 2-D HMMs. We propose a new 2-D HMM-like structure obtained by embedding states within regions of a deformable template structure. With this state-embedded deformable template (SEDT), each region of a deformable template has an underlying observation probability distribution. This structure allows for computation of the P[image/template]. The template that maximizes this probability provides an optimal segmentation of the image. This segmentation capability is demonstrated in facial analysis applications
Keywords :
face recognition; feature extraction; hidden Markov models; image recognition; image segmentation; probability; statistical analysis; 2D HMM; exponential complexity; facial analysis applications; hidden Markov models; merging; observation probability distribution; optimal image segmentation; speech analysis; speech signals; state embedded deformable template; statistical analysis; Deformable models; Face recognition; Head; Hidden Markov models; Image analysis; Image segmentation; Lips; Merging; Speech analysis; Speech recognition;
Conference_Titel :
Image Processing, 1995. Proceedings., International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-8186-7310-9
DOI :
10.1109/ICIP.1995.537695
