By means of the determinantal formulae for inverse and reciprocal differences with coincident data points, the limiting case of Thiele's interpolating continued fraction expansion is studied in this paper and give...By means of the determinantal formulae for inverse and reciprocal differences with coincident data points, the limiting case of Thiele's interpolating continued fraction expansion is studied in this paper and given numerical example shows that the limiting Thiele's continued fraction expansion can be determined once for all instead of carrying out computations for each step to obtain each convergent as done in [3].展开更多
H.264是最新的视频压缩标准,帧内预测是其中一个重要组成部分。JVT(Joint Video Team)提供的参考软件采用全搜索算法进行帧内预测,算法复杂度很高,无法满足实时视频的需求。为了降低帧内预测算法的复杂度,提高编码效率,根据灰度共生矩...H.264是最新的视频压缩标准,帧内预测是其中一个重要组成部分。JVT(Joint Video Team)提供的参考软件采用全搜索算法进行帧内预测,算法复杂度很高,无法满足实时视频的需求。为了降低帧内预测算法的复杂度,提高编码效率,根据灰度共生矩阵及其逆差距提取宏块的纹理特征,对宏块的平坦程度和纹理方向进行预判,提出了一种新的快速帧内预测算法。基于参考软件JM13.2的实验结果表明本算法与全搜索算法相比在信噪比和码率变化极小的情况下,编码速度有显著提高。展开更多
基金Supported by the Foundation for Excellent Young Teachers of the Ministry of Education of China and inpart by the Foundation f
文摘By means of the determinantal formulae for inverse and reciprocal differences with coincident data points, the limiting case of Thiele's interpolating continued fraction expansion is studied in this paper and given numerical example shows that the limiting Thiele's continued fraction expansion can be determined once for all instead of carrying out computations for each step to obtain each convergent as done in [3].
文摘H.264是最新的视频压缩标准,帧内预测是其中一个重要组成部分。JVT(Joint Video Team)提供的参考软件采用全搜索算法进行帧内预测,算法复杂度很高,无法满足实时视频的需求。为了降低帧内预测算法的复杂度,提高编码效率,根据灰度共生矩阵及其逆差距提取宏块的纹理特征,对宏块的平坦程度和纹理方向进行预判,提出了一种新的快速帧内预测算法。基于参考软件JM13.2的实验结果表明本算法与全搜索算法相比在信噪比和码率变化极小的情况下,编码速度有显著提高。
