Hardware implementation and analysis of low power full search block motion estimation techniques

Abstract

Efficient Video coding techniques to reduce the data size of video files has become a necessity in the era of innumerable multimedia applications The video files are compressed at the transmitted side by application of video coding algorithms and decompressed at the receiver side to generate the original video content. High Efficiency Video Coding HEVC newlinealso known as H point 265 is the most successful video compression standard which is extended from Hpoint264 MPEGhypen4 Advanced Video Coding AVC HEVC employs a quad-tree structure in Motion Estimation ME and improved the video quality with high bit rates. In video coding Motion Estimation ME is an essential module which plays a major role in power consumption of video codec ME is determined from the Motion Vector MV Motion Vector is the key element in the motion estimation process It is used to represent a newlinemacro block in a picture based on the position of this macro block or a similar one in another picture, called the reference picture. Motion Vector is determined by the least SAD u,v Sum of Absolute Difference for all possible displacements u,v within a search area Every Motion Vector should be coded and transmitted The power optimizations of ME design is quite typical for video coding applications Full Search Motion Estimation newlineFSME process can essentially advance the video encoder Compression Ratio CR while preserving high quality videos however it is quite expensive and it comprises more than 45 percentage of the complete ME technique The Consecutive Rate Constrained Elimination CRCE method carefully removes contender Motion Vectors while saving the best contender Motion Vector chosen by the Full Search Block Motion Estimation FSBME technique But these techniques failed to consume less power. newline newline newline