Fractional Interpolation Filtering for HEVC Algorithm

Abstract

Video compression is a combination of image compression and motion compensation. Compressing a video refers to decreasing the amount of data used to represent video. Each video contains of numerous frames which are 2D arrays of pixels. Each frame is regarded as a still image. The process of generating a high resolution image from its low-resolution counterpart is referred to as Image interpolation. Hence this process of interpolation can estimate all the pixel values that are required in a frame. But when the frames are vastly uncorrelated, this process becomes inefficient. One of the important ways for video compression is using motion compression vector. The video coding standards has widely used the motion compensation with fractional motion vector. Fractional interpolation filter produces the fractional samples. That is used is an end to end invertibility driven training scheme. Improvement in-video coding efficiency for low bit rate requirement is possible. Video can be shared in less time without changing its quality.