A fast and efficient coding unit size decision algorithm based on temporal and spatial correlation
A fast and efficient coding unit size decision algorithm based on temporal and spatial correlation
复制标题
DOI：
10.1016/j.ijleo.2015.07.026
复制
发表时间：
2015-11
期刊：
影响因子：
3.1
通讯作者：
Qiuwen Zhang;Zhao Jinchao;Xinpeng Huang;Yong Gan
中科院分区：
文献类型：
--
作者：
Qiuwen Zhang;Zhao Jinchao;Xinpeng Huang;Yong Gan
最新的视频编码标准高效视频编码（HEVC）采用递归四叉树结构的编码单元（CU），在编码效率方面显著优于先前的标准。在HEVC的运动估计（ME）过程中，所有可能的CU大小（深度级别）和预测模式都被采用以使用拉格朗日乘子来选择具有最小率失真（RD）成本的一个。这种技术实现了最高的编码效率，但它带来了极高的计算复杂度，限制了HEVC的实际应用。为了降低HEVC的计算复杂度，提出了一种基于时间和空间相关性的快速编码单元大小确定算法。该算法的基本思想是利用视频编码信息在前一帧的时间和空间属性来预测当前的树块预测模式，并提前跳过不必要的CU大小（深度级别）。实验结果表明，该算法可以显着降低计算复杂度的HEVC编码，而保持几乎相同的率失真（RD）性能。
The latest video coding standard high efficiency video coding (HEVC) adopts the recursive quad-tree structured coding unit (CU) significantly outperforms previous standards in terms of coding efficiency. In the motion estimation (ME) process of HEVC, all the possible CU size (depth level) and prediction modes are both employed to select the one with the least rate distortion (RD) cost using Lagrange multiplier. This technique achieves the highest possible coding efficiency, but it brings extremely high computational complexity which limits HEVC from practical applications. In this paper, a fast CU size decision algorithm based on temporal and spatial correlation is proposed to reduce HEVC computational complexity. The basic idea of the algorithm is to utilize video temporal and spatial property of coding information in previous coded frames to predict the current treeblock prediction mode and early skip unnecessary CU size (depth level). Experimental results demonstrate that the proposed algorithm can significantly reduce computational complexity of HEVC encoding while maintaining almost the same rate-distortion (RD) performance.