Note: This tutorial is a record of teacher Jia Zhigang's opencv course study. I would like to express my thanks to teacher Jia.
Dense optical flow tracking is to compare all pixels of the current frame with the previous frame and mark the changes. There are many contrast points, not the characteristic points of contrast changes. So the speed is slower. No sparse optical flow is fast. But sometimes the effect is better than sparse optical flow.
#include <opencv2/opencv.hpp>
#include <iostream>
#include <math.h>
using namespace cv;
using namespace std;
void drawOpticalFlowHF(const Mat &flowdata, Mat& image, int step);
int main(int argc, char** argv)
{
VideoCapture capture;
capture.open("/home/fuhong/code/cpp/opencv_learning/src/object_tracing/video/video_006.mp4");
if (!capture.isOpened()) {
printf("could not load image...\n");
return -1;
}
Mat frame, gray;
Mat prev_frame, prev_gray;
Mat flowResult, flowdata;
capture.read(frame);
cvtColor(frame, prev_gray, COLOR_BGR2GRAY);
namedWindow("flow", CV_WINDOW_AUTOSIZE);
namedWindow("input", CV_WINDOW_AUTOSIZE);
while(capture.read(frame))
{
cvtColor(frame, gray, COLOR_BGR2GRAY);
if (!prev_gray.empty()) {
calcOpticalFlowFarneback(prev_gray, gray, flowdata, 0.5, 3, 15, 3, 5, 1.2, 0);
cvtColor(prev_gray, flowResult, COLOR_GRAY2BGR);
drawOpticalFlowHF(flowdata, flowResult, 10);
imshow("flow", flowResult);
imshow("input", frame);
}
char c = waitKey(1);
if (c == 27) {
break;
}
}
return 0;
}
void drawOpticalFlowHF(const Mat &flowdata, Mat& image, int step) {
for (int row = 0; row < image.rows; row++) {
for (int col = 0; col < image.cols; col++) {
const Point2f fxy = flowdata.at<Point2f>(row, col);
if (fxy.x > 1 || fxy.y > 1) {
line(image, Point(col, row), Point(cvRound(col + fxy.x), cvRound(row + fxy.y)), Scalar(0, 255, 0), 2, 8, 0);
circle(image, Point(col, row), 2, Scalar(0, 0, 255), -1);
}
}
}
}
The effect is as follows:
