一种低复杂度运动补偿时间滤波新策略研究

Research on a new scheme of low-complexity motion compensated temporal filtering

针对传统运动补偿时间滤波(MCTF)结构存在着编码速度慢的缺点,提出了一种新的基于低复杂度的运动补偿时间滤波结构。此结构实现了对传统MCTF结构的优化,并精简了传统MCTF的计算步骤,同时给出了一种运动向量多重预测的方法,并通过连续帧间运动向量预测与运动向量多重预测门限值参数AE的选择,实现了对编码计算复杂度的控制。实验数据表明,此结构编码时间比传统MCTF结构缩短约12%。

Aiming at the shortcoming of traditional MCTF structure with slow coding speed, a new structure of low-complexity motion-compensated temporal filtering is proposed. This structure can optimize the traditional MCTF structure, simplify the calculating steps of traditional MCTF structure, and provide a method of motion vector muhi-prediction. Moreover, it can control the coding computational complexity through selecting the gate-valueAEin the continuous inter-layer motion vector prediction and motion vector multi-prediction. The experimental data shows that the coding time is reduced by 12% compared with the traditional MCTF structure.