Image partitioning for fractal coding under rate and complexity constraints

Fractal image coding is a relatively new technique for compact representation of an image by exploiting self-similarities between parts of the image and other parts in it at a higher resolution. The various parts at different resolutions are consequences of a partition grid and splitting criterion applied to the image. We propose a partitioning criterion which takes in consideration the computational complexity of the encoding process. Based on this criterion we develop an algorithm for adaptive image partitioning achieving a range of rates under a computational complexity constraint. The proposed algorithm results in a reduction of the computational complexity as compared to other known algorithms at the same rate-distortion operating point.