Paper Title
Machine Learning Based Algorithm For High Efficiency Video Coding

Abstract
High Efficiency Video Coding HEVC/H.265 is a futuristic standard for video compression with the promise of video quality standards at approximately 50% lower bit rates. Over the past decade, MPEG and H.264/Advanced Video Coding were standards used for compression widely in various multimedia applications. They provided extensive support for a variety of image and video resolutions with comfortable compression standards. However, with the increasing demand for high quality video resolutions like High Definition/Ultra High Definition (HD/UHD) these standards could not provide comparable compression efficiency. HEVC due to its increased flexibility in hybrid coding structures is suitable to accommodate high quality content like HD/UHD. However, this increases the complexity of computations up by 10 times as compared to H.264/AVC. The computations are mainly for selection of appropriate Coding Units (CU) in order to view the video content without compromising on quality. Machine learning is a viable solution for coding unit selection as it can learn/analyze patterns through extracted features from any data set. In this paper, a machine learning based algorithm using Support Vector Machine (SVM) is proposed with an attempt to select the appropriate coding units and to eliminate unnecessary computations. Index Terms- AVC, Coding Unit (CU), HEVC, machine learning.