【摘要】 提出一种新的改进三步搜索法(NITSS),该方法充分利用视频序列运动矢量在空间分布上的中心偏置特性,在三步搜索算法的基础上引入了以“×”字型分布的5个点构成的搜索点群,并通过同一帧相邻宏块的运动矢量之间存在的空间相关性来预测初始运动矢量,用来加强中心偏置特性。仿真结果表明,同TSS算法相比,NITSS算法降低了搜索运算量,提高了搜索精度。更多还原

【Abstract】 In this paper, we proposed a new improved three-step search algorithm(NITSS). The features of NITSS are that it employs a center-biased checking point pattern in the first step, which is derived by making the search adaptive to the motion vector distribution, we utilized the spatial correlation of the adjacent macroblocks to predict the initial motion vector, which can enhence the center-biased distribution characteritic. Simulation results show that, as compared to TSS, NITSS reduces the complexity and improves the precision in the process of search.更多还原