A new motion compensated predictive coding method for an image sequence at a low bit rate is presented. Motion compensation is based on the segmentation of the reconstructed image performed in both encoder and decoder. This will eliminate the need to transmit the region shape information. Also, motion vector prediction is implemented to reduce the overhead for motion information. Motion-compensated prediction errors are transformed using the discrete cosine transform, and the coefficients are quantized and entropy coded as recommended by CCITT. Computer simulation shows that the proposed coding algorithm significantly reduces the block distortion, as conventional block matching algorithms do at very low bit rates (e.g., 0.02 bit/pixel).