Low complexity video compression algorithm using AMBTC

Abstract

A computationally simple algorithm for video compression is presented in this paper. The algorithm gives excellent quality image sequence reconstruction at a peak-signal-to-noise-ratio (PSNR) of approximately 37 decibels (dB) and a very low bit rate of 0.2 bits-per-pixel (bpp). The scheme utilizes absolute moment block truncation coding (AMBTC) for intraframe coding, in conjunction with three-step algorithm for block based motion compensation, and a simple quantization scheme for interframe error coding. The procedural steps in the algorithm are very similar to the MPEG standard, which can facilitate the implementation of the proposed algorithm in the existing infrastructure. The results of the simulations performed on three different image sequences depicting different amounts of motion and background activity are presented. The qualitative and quantitative comparisons of the proposed algorithm, with an existing video compression standard and with other similar research are presented. The significant advantage gained in terms of computational complexity and the probable use of the algorithm for progressive transmission of video is illustrated. The modifications of the basic algorithm are also presented in this paper.