Fractal coding of image sequence using extended circular prediction mapping

This paper proposes a novel algorithm for fractal coding of color image sequence, based on the extended CPM (Circular Prediction Mapping). In the extended CPM, each range block is approximated by a domain block in the adjacent frame, which is of the same size as the range block. Therefore the proposed domain-range mapping is similar to the block matching algorithm in the motion compensation techniques, and we can exploit the temporal correlation in moving image sequence effectively. Also we show that fast decoding is possible, since the decoder requires about 1 multiplication and 3 additions per pixel for each Y, U, V components. The computer simulation results on real image sequences demonstrate that the proposed algorithm provides very promising performance at low bit-rate.