在H.S.Kang的ME Skip运动估计算法的基础上,提出了一种改进的ME Skip运动估计算法.新算法对运动估计匹配误差差分绝对值之和(SAD)的阈值设置作了改进.新算法设置了两个SAD阈值,较小的SAD阈值用以克服误差传播,较大的SAD阈值用以尽可能的避免运动估计.当基于预测运动矢量的匹配误差介于两SAD阈值之间时,用小窗口搜索代替运动估计.实验结果表明在编码效率略有提高的同时,新算法进一步减少了运动估计运算量的10%～20%. On the basis of HSKang ME Skip motion estimation algorithm, an improved ME Skip motion estimation algorithm is proposed. The new algorithm improves the threshold setting of sum of difference absolute difference (SAD) of motion estimation matching error. Two SAD thresholds are used, the smaller SAD threshold is used to overcome the error propagation, and the larger SAD threshold is used to avoid motion estimation as much as possible.When the matching error based on the predicted motion vector is between two SAD thresholds, Window search instead of motion estimation.The experimental results show that while the coding efficiency is slightly improved, the new algorithm further reduces the motion estimation computation by 10% ~ 20%.