In this paper, we propose modifications to a H.263 codec for cod-ing very low bit-rate head and shoulder video sequences. The im-portant facial features (eyes, nose and mouth) are tracked from frame to frame and the knowledge of their positions is used to im-prove quantization in these regions of interest, compared to the other facial regions and the background, which have coarse quant-ization.
... [Show full abstract] Results are presented which show the effectiveness of the method using common head and shoulder video sequences.