论文部分内容阅读
根据质量可伸缩视频编码(SVC)的特征,提出一种适合于粗粒度质量SVC增强层的帧间快速编码算法。首先,根据基本层宏块的模式和率失真值,预测当前宏块编码模式采用的顺序;然后,根据残差系数的分布特征提出了残差系数判断,根据空间相关性提出了率失真值判断;最后,根据模式已预测的顺序编码当前宏块,利用残差系数判断和率失真值判断进行提前终止,以提高编码速度。实验结果表明,与标准算法相比,本文算法的计算复杂度平均降低了68%,同时对编码质量和比特率几乎没有影响。
According to the characteristics of Scalable Scalable Video Coding (SVC), a fast inter-frame coding algorithm is proposed for coarse-grained SVC enhancement layer. First of all, the order of the current macroblock coding modes is predicted according to the mode and rate-distortion of the base layer macroblocks. Then, according to the distribution characteristics of the residual coefficients, the coefficient of residuals is proposed. According to the spatial correlation, Finally, the current macroblock is coded according to the predicted sequence of the mode, and pre-terminated by using the residual coefficient judgment and the rate distortion value judgment to improve the coding speed. Experimental results show that compared with the standard algorithm, the computational complexity of this algorithm is reduced by 68% on average, while it has almost no effect on the coding quality and bit rate.