An approach to statistical lip modelling for speaker identification via chromatic feature extraction

Author

Wark, T. ; Sridharan, S. ; Chandran, V.

Author_Institution

Sch. of Electr. & Electron. Syst. Eng., Queensland Univ. of Technol., Brisbane, Qld., Australia

Volume

1

fYear

1998

fDate

16-20 Aug 1998

Firstpage

123

Abstract

This paper presents a novel technique for the tracking of moving lips for the purpose of speaker identification. In our system, a model of the lip contour is formed directly from chromatic information in the lip region. Iterative refinement of contour point estimates is not required. Colour features are extracted from the lips via concatenated profiles taken around the lip contour. Reduction of order in lip features is obtained via principal component analysis (PCA) followed by linear discriminant analysis (LDA). Statistical speaker models are built from the lip features based on the Gaussian mixture model (GMM). Identification experiments performed on the M2VTS¹ database, show encouraging results

Keywords

Gaussian distribution; biometrics (access control); feature extraction; image recognition; speaker recognition; statistical analysis; tracking; GMM; Gaussian mixture model; LDA; M2VTS¹ database; PCA; chromatic feature extraction; concatenated profiles; linear discriminant analysis; lip contour model; principal component analysis; speaker identification; statistical lip modelling; statistical speaker models; tracking; Active shape model; Data mining; Feature extraction; Laboratories; Linear discriminant analysis; Lips; Polynomials; Principal component analysis; Speech; Systems engineering and theory;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on

Conference_Location

Brisbane, Qld.

ISSN

1051-4651

Print_ISBN

0-8186-8512-3

Type

conf

DOI

10.1109/ICPR.1998.711095

Filename

711095