Lip features automatic extraction

Author

Lievin, M. ; Luthon, F.

Author_Institution

Signal & Image Lab., Grenoble Nat. Polytech. Inst., France

fYear

1998

fDate

4-7 Oct 1998

Firstpage

168

Abstract

An algorithm for speaker´s lip segmentation and features extraction is presented. A color video sequence of speaker´s face is acquired, under natural lighting conditions and without any particular make-up. First, a logarithmic color transform is performed from the RGB to HI (hue, intensity) color space. Second, a statistical approach using Markov random field modeling determines the red hue prevailing region and motion in a spatiotemporal neighborhood. Third, the final label field is used to extract ROI (region of interest) and geometrical features

Keywords

Markov processes; feature extraction; image colour analysis; image segmentation; lighting; random processes; statistical analysis; video signal processing; LIP-MAD algorithm; Markov random field modeling; RGB color space; ROI extraction; color video sequence; face; geometrical features; hue-intensity color space; label field; lip features automatic extraction; lip segmentation; logarithmic color transform; natural lighting conditions; red hue prevailing region; region of interest; spatiotemporal neighborhood motion; statistical approach; Automatic speech recognition; Data mining; Feature extraction; Humans; Image processing; Image segmentation; Lighting; Markov random fields; Mouth; Pixel; Spatiotemporal phenomena; Video sequences;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 1998. ICIP 98. Proceedings. 1998 International Conference on

Conference_Location

Chicago, IL

Print_ISBN

0-8186-8821-1

Type

conf

DOI

10.1109/ICIP.1998.727160

Filename

727160