输入两幅图像，从中提取好跟踪的sift点，计算相邻帧的sift点的运动矢量-measure the motion vector of sift points in two images with light flow method.

利用openCV，首先得到图像中的强边界作为跟踪的特征点，调用函数，输入两幅连续的图像，并在第一幅图像里选择一组特征点，输出为这组点在下一幅图像中的位置。再把结果过滤，去掉不好的特征点。把特征点的跟踪路径标示出来。(Using openCV, we use goodFeaturesToTrack function to get strong edges in the image as the feature point tracking, next to call calcOpticalFlow

光流法实现对视频目标的提取，适用于运动相机拍摄的物体(Optical flow method to achieve the extraction of video objects, suitable for sports camera shooting objects)