Computational Complexity Scalable ME Algorithm and Architecture

A computational complexity scalable block-matching algorithm (CCS-BMA) is presented in this paper for the motion estimation of video encoding. The CCS-BMA shares the feature of fast convergence in TSS and the advantage of center bias searching in FSS and DS. With the uniform shape of searching pattern, the VLSI implementation of the CCS-BMA is more convenient than that of other fast motion estimation algorithms. The scalability of computational payload can be achieved through searching steps adjustment and block pixel sub-sampled. The proposed algorithm and architecture are suitable for low power implementation of video encoding that needs flexible scalable capability