A New Manifold Representation for Visual Speech Recognition

Author

Yu, Dahai ; Ghita, Ovidiu ; Sutherland, Alistair ; Whelan, Paul F.

Author_Institution

Dublin City Univ., Dublin

fYear

2007

fDate

5-7 Sept. 2007

Firstpage

210

Lastpage

210

Abstract

In this paper, we propose a new manifold representation for visual speech recognition. The developed system consists of three main steps: a) lip extraction from input video data, b) generate the expectation-maximization PCA (EMPCA) manifolds for the entire image sequence and perform manifold interpolation and re-sampling, c) classify the manifolds using a HMM classifier to identify the words described by the lips motions in the input video sequence.

Keywords

expectation-maximisation algorithm; face recognition; feature extraction; hidden Markov models; image classification; image motion analysis; image sequences; interpolation; principal component analysis; speech recognition; video signal processing; HMM classifier; expectation-maximization PCA manifold; image sequence; lip motion extraction; manifold interpolation; manifold representation; manifold resampling; visual speech recognition; Computer vision; Hidden Markov models; Image processing; Interpolation; Lips; Machine vision; Manifolds; Speech recognition; Systems engineering and theory; Video sequences;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Vision and Image Processing Conference, 2007. IMVIP 2007. International

Conference_Location

Kildare

Print_ISBN

978-0-7695-2887-8

Type

conf

DOI

10.1109/IMVIP.2007.35

Filename

4318164