基于SSE2对H.265/HEVC中分像素插值的优化

摘要

Sub-pixel interpolation algorithm is a key technology in video coding algorithm, which has important influence on video coding performance. In the latest video coding standard H.265/HEVC, interpolation filters based on DCT are used to interpolate half pixel and quarter pixel, but it has very high computation complexity, especially for high definition and ultra-high resolution video coding applications. This paper introduces the background of DCT-based interpolation filters in detail, and then optimizes its implementation by SSE2 instruction set. Experimental results demonstrate that the proposed method improves the computation efficiency by 217%~259%as compared with the reference implementation in HM software.%分像素插值算法是视频编码算法中的一项关键技术，对视频编码性能有非常重要的影响。在最新的视频编码标准H.265/HEVC中，提出了使用DCT插值滤波器（DCTIF）分别进行半像素和1/4像素插值，但是计算复杂度较大，尤其对于高分辨率和超高分辨率的视频编码应用。对滤波器系数的理论推导过程进行了详细介绍，并利用SSE2指令集进行了优化设计和实现，实验结果表明该方法比参考代码中的方法提高了217%~259%的性能。