For semantic coding of videophone sequences at very low bit rates,the adaptation of a face model to automatically estimate facial featuresis necessary. In this contribution, an algorithm for the automaticestimation of the chin and cheek contours of a person is presented. Thechin and cheeks are represented by a deformable template consisting ofparabolas. Cost functions are established and minimized to find the bestfit of the template to the chin and cheek contours of the person. In allexamined frames of the videophone sequences “Akiyo” and“Claire”, the manually determined real contour of the chinand the cheeks is approximated by the estimated contours with goodaccuracy. In more than 60% of all frames, no deviation between the realcontours and the estimated contours occurs
展开▼
