基于运动相关的运动估计自适应快速算法

A Spatial-temporal Correlation-based Self-adaptive Fast Searching Al gorithm for Motion Estimation

根据视频运动的空间和时间相关性,以当前帧相邻块和前一帧重建VOP对应位置块及其相邻块为基础,并且自适应地调整当前帧和前一帧各块对初始搜索点确定的权值,完成初始搜索点的准确预测,然后针对视频的不同类型选择搜索策略,结合自适应调整的中止SAD阀值,适时中止搜索的过程。模拟实验表明,这种快速算法对初始搜索点的预测准确,搜索策略应用得当,极大地减少了运算量,各项性能指标都非常接近全搜索算法。

According to the spatial and temporal correlation of the video motion,this paper proposes a method for accurate prediction of initial point and self-adaptive adjustment of the weight between current and previous frame on the basis of adjacent blocks at current frame ,correlated block and adjacent blocks of reconstructed VOP at previous frame ,and then applies different search strategy to different motion activities together with self-adaptive adjustment of SAD threshold to stop searching timely.Simulation results show that this fast search algorithm predicts the initial point and applies search strategy efficiently.As a result,it reduces the computational load greatly and obtains almost the same performance as full-search algorithm.