Rate Model Considering Nontexture Bits for High Efficiency Video Coding

Abstract

One of the challenges in video rate control lies in accurately achieving the target bitrate by determining coding parameter set. In HEVC, nontexture bits become more significantly important, due to the adoption of more coding tools compared to H. 264 standards. In this paper, we propose an accurate rate model to estimate the size of nontexture bits accurately. In addition, parameter update conditions and a model failure detection scheme are introduced to avoid large deviations of the frame size from the target size. In comparison to the original rate model without nontexture bits estimation, the proposed method can achieve better video quality with a PSNR gain up to 0.47 dB and reduce the mismatch between actual frame bits and target ones by up to 64.3%.