This paper describes the work carried on by the ACTS project VIDAS1 in the area of facial image analysis for the synthesis of 3D animated virtual faces. The proposed facial decoder is compliant to MPEG-4 standard and is capable of reproducing the characteristics of any real specific face and guaranteeing synchronisation of lip movements with speech. Evidence of the achievable performance is reported at the end of the paper by means of pictures showing the capability of the system to reshape its geometry according to the decoded MPEG-4 facial calibration parameters and its effectiveness in performing facial expressions. A variety of possible applications is proposed and discussed.