Improved Bit Allocation Using Distortion for the CTU-Level Rate Control in HEVC

Abstract

The research on rate control has always been a hot topic in video coding. Although H.265/HEVC is the latest video coding standard, the mismatch between the number of target bits and that of actual coding bits in the coding process still exists. A large number of experimental results have shown that the R-D model is superior to the R-λ model in terms of the matching degree of bit allocation. In this paper, a novel bit allocation method is proposed. First, the recursive Taylor expansion equation solving method is used to solve the rate-distortion (constrained problem) optimization equation, and then the R-D model is used to optimize the CTU-level target bit allocation.The simulation results indicate that the average bitrate error is only 0.19%, while the mean peak signal-to-noise ratio (MPSNR) increases by 0.14 dB.