TOWARDS AN AUDIOVISUAL VIRTUAL TALKING HEAD: 3D ARTICULATORY MODELING OF TONGUE, LIPS AND FACE BASED ON MRI AND VIDEO IMAGES

A linear three-dimensional articulatory model of tongue, lips and face is presented. The model is based on a linear component analysis of the 3D coordinates defining the geometry of the different organs, obtained from Magnetic Resonance Imaging of the tongue, and from front and profile video images of the subject’s face marked with small beads. In addition to a common jaw height parameter, the tongue is controlled by five parameters while the lip and face are driven by four parameters, that can be interpreted in phonetic / articulatory terms. This model has been finally integrated into the ICP virtual talking head. 1