next up previous contents
Next: Initialisation of the Feature Up: Discussion of the Proposed Previous: Discussion of the Proposed

Localisation of the Mouth

In this section we describe the means by which the lips of the speaker are located on a frame by frame basis. The process described below involves an initialisation stage, whereby the facial feature templates are extracted, and a tracking stage, which makes use of OTC to determine the mouth location with respect to the eyes and nose.





Markus Weber
Tue Jan 7 15:44:13 PST 1997