Face feature detection and model design for 2D scalable model-based video coding

This paper addresses our proposed methods for face feature detection and 2D scalable face model design in order to achieve scalable model-based video coding. Firstly, fast and reliable algorithms have been proposed for eye, mouth and chin detection. Then, based on the detected face features and human face muscular distribution, a heuristic scalable face model is designed. Experimental results show that this scalable face model can represent face motion precisely.