A Fast Intra Prediction Decision Algorithm for HEVC based on Image Texture Complexity
Hang Zhuang, Dan Fan, Liang Liu, Rui Wang, Changhao Sui · 2021
High Efficiency Video Coding (HEVC) shows excellent performance in improving video compression rate, but it also has considerable computational complexity. In the process of intra prediction, Coding Unit(CU) size decision and Prediction Unit(PU) prediction mode selection consume a lot of time. Aiming at these two aspects, this paper uses Sobel operator to calculate the luma gradient and texture complexity of the image. Then we determine the split methods of some CUs in advance according to the texture complexity, and optimize the PU prediction mode selection process. Experimental results show that the proposed algorithm can reduce the computation time by 49.39% compared with the algorithm in HM16.9 platform, increase the bit rate by 1.08%, and decrease the PSNR by 0.06dB, which has no impact on the video quality.