复杂度可伸缩的帧间模式选择

Inter mode decision of complexity scalability

针对H.264/AVC中复杂度无法满足终端设备异构性的问题,提出一种新的帧间模式选择算法。利用相邻宏块的空间相关性和时间相关性,进行模式预测,并通过复杂度伸缩因子,控制参与预测的模式的数目,使复杂度在20%到100%之间灵活变化。实验结果表明,该算法能在视频质量和复杂度之间形成良好的折中,适应从高端设备到低端设备的计算能力差异。

In order to resolve the problem that computation complexity of H.264/AVC can not satisfy the diversity of computing resources of terminal equipment,the authors proposed a novel algorithm of inter mode decision.The algorithm utilized the spatial correlation and time correlation of adjacent macroblocks to predict the modes of current macroblocks,and controlled the number of modes involved in the prediction by scalability factor;hence the complexity can change flexibly from 20% to 100%.The experimental results show that the proposed method can achieve the tradeoff between the accuracy of motion estimation and its computational complexity,fit for the different computing power of different equipments.