Very low bit-rate video coding based on a method of facial area specification

Abstract

An efficient video coding algorithm for coding the upper half of a human body is proposed. In the coding system, the facial area, which is a moving object in the image sequence, is extracted from both the sending and receiving side without any control information. The candidate areas of the person are segmented using color information and the moving areas. The color information, which represent skin color, is extracted with a modified HSV color system. The moving areas are estimated using the motion parameters of objects, which are utilized for constructing the previous frame. The face is then extracted by merging the candidate areas based on the hue element of the modified HSV color system. The standard "Claire" video is used for the computer simulation, and facial areas are accurately extracted in all frames. Furthermore, to improve the subjective evaluation, the eyes and mouth are extracted as the most important areas in the face and coded precisely.