The training of Karhunen–Loève transform matrix and its application for H.264 intra coding

The training of Karhunen–Loève transform matrix and its application for H.264 intra coding
复制标题

KarhunenâLoève变换矩阵的训练及其在H.264帧内编码中的应用

DOI:
10.1007/s11042-008-0221-4
复制
发表时间:
2008
影响因子:
3.6
通讯作者:
Jiazhong Chen
Jiazhong Chen
中科院分区:
计算机科学4区
文献类型:
--
作者:
Yi Gao;Jingli Zhou;Lai-Man Po;Shengsheng Yu;Jiazhong Chen

文献摘要

参考文献

相似文献

在H.264/AVC中,对帧内预测后的残差信号进行4×4离散余弦变换(DCT)去相关。实际上,具有不同预测模式的残差块表现出不同的频率特性。因此,固定的变换矩阵不能很好地匹配残差信号的能量分布,从而降低了去相关性能。幸运的是,相同模式的残差块的能量分布相对一致,这使得为每种模式训练普遍良好的卡尔胡宁-洛伊夫变换(KLT)矩阵成为可能。本文提出了一种最优频率匹配(OFM)算法来训练剩余块的KLT矩阵,并训练了9个对应于4×4帧内块的9种预测模式的KLT矩阵。实验结果表明,与采用4×4离散余弦变换的H.264相比,具有训练矩阵的KLT具有持久的增益,平均峰值信噪比(PSNR)提高了0.22dB,最大提高了0.33dB。
In H.264/AVC, 4 × 4 discrete cosine transform (DCT) is performed on the residual signals after intra prediction for decorrelation. Actually, residual blocks with different prediction modes exhibit different frequency characteristics. Therefore, the fixed transform matrix cannot match the energetic distribution of residual signals very well, which degrades the decorrelation performance. Fortunately, the energetic distributions of residual blocks with the same mode are relatively coincident, which makes it possible to train a universally good Karhunen–Loève transform (KLT) matrix for each mode. In this paper, an optimal frequency matching (OFM) algorithm is proposed to train KLT matrices for residual blocks and nine KLT matrices corresponding to nine prediction modes of 4 × 4 intra blocks are trained. Experimental results show that KLT with trained matrices yields a persistent gain over H.264 using 4 × 4 DCT with an average peak signal-to-noise ratio (PSNR) enhancement of 0.22dB and a maximum enhancement of 0.33dB.
DOI: 10.1109/83.791958
发表时间: 1999-10
期刊: IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
影响因子: --
作者:
M. Effros;P. Chou;R. Gray
通讯作者: M. Effros;P. Chou;R. Gray
DOI: 10.1109/mcom.2006.1678121
发表时间: 2006-08
影响因子: 11.2
作者:
D. Marpe;T. Wiegand;G. Sullivan
通讯作者: D. Marpe;T. Wiegand;G. Sullivan
DOI: 10.4218/etrij.06.0206.0095
发表时间: 2006-10
期刊: ETRI Journal
影响因子: 1.4
作者:
Yung-Lyul Lee;Ki-Hun Han;D. Sim;Jeongil Seo
通讯作者: Yung-Lyul Lee;Ki-Hun Han;D. Sim;Jeongil Seo
DOI: 10.4218/etrij.05.0905.0032
发表时间: 2005-10
期刊: ETRI Journal
影响因子: 1.4
作者:
Kibum Suh;Seongmo Park;Hanjin Cho
通讯作者: Kibum Suh;Seongmo Park;Hanjin Cho