A Hierarchical Segmentation Algorithm for Face Analysis. Application to Lipreading

Marc Liévin (1) , Franck Luthon (2)

1 LIS - Laboratoire des images et des signaux
2 T2I

A hierarchical algorithm for face analysis is presented in this paper. A color video sequence of speaker's face is acquired, under natural lighting conditions and without any particular make-up. The application aims at providing geometrical features of the face for scalable video transmission when no specific model of the speaker face is assumed. First, a logarithmic hue transform is performed from RGB to HI (hue, intensity) color space. Next, a Markov random field modeling regularizes motion and hue information within a spatiotemporal neighborhood. The hierarchical segmentation labels the different areas of the face. Results are shown on the lower part of the face and compared with standard color segmentation algorithm (fuzzy c-means). A speaker's lip shape with inner and outer borders is extracted from the final labeling and used to initialize an active contours stage.

Domaines: Traitement du signal et de l'image [eess.SP]

Marc Liévin, Franck Luthon. A Hierarchical Segmentation Algorithm for Face Analysis. Application to Lipreading. IEEE Int. Conf. on Multimedia and Expo (ICME'2000), Jul 2000, New-York, United States. pp.1085-1088. ⟨hal-00960991⟩