HEVC快速算法文献综述

FastCoding Unit Size Selection for HEVC based on Bayesian Decision Rule

 

Abstract—HighEfficiency Video Coding (HEVC) is being developed by JointCollaborative Team on Video Coding (JCTVC).The new standardaims at a significant improvement on Rate Distortion (RD)performance over state-of-the-art video coding standards.Some new efficient coding tools are adopted,such as flexibledata structure representation: Coding Unit(CU), PredictionUnit (PU), and Transform Unit (TU).Meanwhile, these newtools impose lots of complexity. We propose a CU sizeision algorithm that greatly reduces the complexity of HEVC.The proposed algorithm collects relevant andcomputational-friendly features to assist decisions on CU splitting. Tominimize the RD loss of the fast algorithm, a Bayesian decisionrule is defined, based on which the encoder can save 41.4% encoding time on average (up to 66%) whilst suffers from negligible loss (1.88%) on RD performance.

 

Fast Coding AlgorithmBased on Adaptive CodingDepth Range Selection for HEVC

 

Abstract—The JCT-VCis developing the next-generation video codingstandard, called high efficiency video coding (HEVC). In HEVC, there are three units in a blockstructure. These units are acoding unit (CU), a prediction unit (PU), and a transform unit (TU). The CU is the basic unit forregion splitting (like amacroblock). The CU performs recursive splitting into four blocks of equal size, starting from thetreeblock. We use mode informationof the current CU to avoid unnecessary CU splitting.To reduce the computationalcomplexity, we propose a depth range selection mechanism (DRSM) using aselecting adaptive depth range.This depth range selection mechanism allows for effective splitting by only processing CUs that need thisprocess.Experimental results show that the proposed algorithm achieves an approximate 48% reduction in encodingtime for Random Access HE,compared to the HEVC test model (HM) 6.0 encoderwith a BD-bitrate loss of 1.2%.

 

 

An effective CU size decisionmethod for HEVC encoders

Abstract:The emerging high efficiency video coding standard (HEVC) adopts thequadtree-structured coding unit (CU).Each CU allows recursive splitting intofour equal sub-CUs. At each depth level (CU size), the test model of HEVC (HM) performs motion estimation (ME) withdifferent sizes including 2N×2N, 2N×N, N×2N, and N×N.ME process in HM is performed using all the possibledepth levels and prediction modes to find the one with the least ratedistortion (RD) cost usingLagrange multiplier. This achieves the highest coding efficiency but requires avery high computational complexity. In this paper, we propose a fast CU size decision algorithm for HM.Since the optimal depth level is highly content-dependent,it is not efficientto use all levels. We can determine CU depth range (including the minimum depthlevel and the maximum depthlevel) and skip some specific depth levels rarely used in the previous frameand neighboring CUs. Besides, the proposed algorithm also introduces early termination methods basedon motion homogeneity checking, RD cost checking and SKIP mode checking to skip ME on unnecessary CU sizes.Experimental results demonstrate that our algorithm cansignificantly reduce complexity while maintainingalmost the same RD performance as the original HEVC encoder.

 

Early Determination of ModeDecision for HEVC

Abstract—In thispaper, we propose a fast decision method scheme to reduceencoder complexity of high efficiency video coding. It is anearly detection of SKIP mode in one CU-level based on thedifferential motion vector (DMV) and coded block flag (CBF).Experimental results show that the encoding complexity can bereduced by up to 34.55% in random access (RA) configurationand 36.48% in low delay (LD) configuration with only a littlebit of rate increment compared to the high efficiency videocoding test model (HM) 4.0 reference software.