简体繁体 English

实时使用opencv的光流

[英]optical flow using opencv in real time

原文 2012-07-20 19:31:32 1 1 c++/ c/ opencv

I am using Pyramid Lukas Kanade function of OpenCV to estimate the optical flow. 我正在使用OpenCV的Pyramid Lukas Kanade功能来估算光流。 i call the cvCalcOpticalFlowPyrLK(const CvArr* prev, const CvArr* curr, CvArr* prev_pyr, CvArr* curr_pyr, const CvPoint2D32f* prev_features, CvPoint2D32f* curr_features, int count, CvSize win_size, int level, char* status, float* track_error, CvTermCriteria criteria, int flags): 我调用cvCalcOpticalFlowPyrLK（const CvArr * prev，const CvArr * curr，CvArr * prev_pyr，CvArr * curr_pyr，const CvPoint2D32f * prev_features，CvPoint2D32f * curr_features，int count，CvSize win_size，int level，char * status，float * track_error，CvTermCriteria criteria，int flags）：

i have a problem in getting the features coordinate. 我在获取要素坐标时遇到问题。 i have a set of 8 coordinates in my CvPoint2D32f* prev_features data and i have to display the future features points on my capturing image. 我在我的CvPoint2D32f * prev_features数据中有一组8个坐标，我必须在我的捕获图像上显示未来的特征点。 my work is in real time, so i have to detect and display the points continues.. hope u can give me some suggestion which i have to consider in this section. 我的工作是实时的，所以我必须检测并显示积分继续..希望你能给我一些建议，我必须在本节考虑。 thanks alot 非常感谢

1 个解决方案

It's hard to say what's wrong with your code, because I don't see it. 很难说你的代码出了什么问题，因为我没有看到它。 Anyway, did you look in default sample? 无论如何，你看过默认样本吗？ I mean lkdemo.cpp . 我的意思是lkdemo.cpp 。 It obviously should help you. 它显然应该对你有帮助。