Homogeneity-based fast CU partitioning algorithm for HEVC intra coding

Abstract

High Efficiency Video Coding (HEVC) is a new video coding standard released as a successor for H.264/AVC. It expected to reduce by 50% the bitrate for the same perceptual quality. One of the major contributors to the higher compression performance of HEVC is the introduction of larger Coding Units (CU) with recursive partitioning mechanisms. This encoding performance is accompanied by a high computational complexity, which makes it very difficult to achieve real-time encoding especially if we aim to implement this encoder on embedded platforms. In this context, this paper suggests a fast CU partitioning algorithm for Intra-only (All Intra) configuration. The proposal aims to early terminate CU partitioning for homogeneous regions in the video frame, or skip some depths for high textured regions. The decision of Split/Non-split is based on homogeneity classification algorithm, which allows us to avoid the test of the all depths in order to determine the best CU size. Experimental results confirm that the proposed approach can reduce up to 41% of encoding time in average for different video classes and can reach up to 58% for high homogenous texture video sequences. Keywords: HEVC coding, CU partitioning, Complexity reduction, Intra coding