Opencv video tracking "suggestions collection"

Hello everyone , I meet you again , I'm your friend, Quan Jun .

What is object tracking ？

In short , Locating objects in consecutive frames of video is called track .

The definition sounds straightforward , But in computer vision and machine learning , Tracking is a very broad term , Cover conceptually similar but technically different ideas . for example , Usually in Object tracking Next, study all the following different but related ideas

1. Dense optical flow ： These algorithms are helpful to estimate the motion vector of each pixel in the video frame .
2. Sparse optical flow ： These algorithms , Such as Kanade-Lucas-Tomashi（KLT） Feature tracker , Track the position of several feature points in the image .
3. Kalman filtering ： A very popular signal processing algorithm , It is used to predict the position of a moving object based on previous motion information . One of the early applications of this algorithm is missile guidance ！ Also mentioned here ,“ Is to guide Apollo 11 The on-board computer of the lunar module landing on the moon has a Kalman filter ”.
4. Meanshift and Camshift： These are algorithms for locating the maximum value of the density function . They are also used to track .
5. Single object tracker ： In this kind of tracker , The first frame is marked with a rectangle , To indicate the location of the object we want to track . Then, the tracking algorithm is used to track the object in subsequent frames . In most practical applications , These trackers are used in combination with object detectors .
6. Multiple object tracking algorithm ： When we have a fast object detector , It is meaningful to detect multiple objects in each frame and then run the tracking search algorithm to identify which rectangle in one frame corresponds to the rectangle in the next frame .

Tracking and detection

If you've ever played OpenCV Face detection , You know it works in real time , You can easily detect the face in each frame . that , Why do you need to track first ？ Let's explore the different reasons why you might want to track objects in video , Not just repeat detection .

1. Tracking is faster than detection ： Generally, the tracking algorithm is faster than the detection algorithm . The reason is simple . When you track objects detected in the previous frame , You know a lot about the appearance of this object . You can also know the position in the previous frame and the direction and speed of its movement . therefore , In the next frame , You can use all this information to predict the position of the object in the next frame , And make a small search around the expected position of the object , To accurately locate the object . A good tracking algorithm will use all its information about the object , The detection algorithm always starts from scratch . therefore , When designing effective systems , Usually every n One operation and one object detection in between n-1 The frame in which the tracking algorithm is used . Why don't we directly detect the object in the first frame and then track ？ exactly , Tracking can benefit from the additional information it has , But when they fall behind obstacles for a long time , Or if they move too fast for the tracking algorithm to catch up , You may also lose tracking of objects . Cumulative errors in tracking algorithms are also common , The bounding box of the tracked object will slowly deviate from the object it is tracking . In order to solve these problems by tracking algorithm , Run the detection algorithm every once in a while . The detection algorithm is trained on a large number of examples of objects . therefore , They know more about the general classes of objects . On the other hand ,
2. When the detection fails , Tracking can help ： If you run a face detector on a video and your face is blocked by an object , Then the face detector is likely to fail . On the other hand , A good tracking algorithm will handle some degree of occlusion . In the video below , You can see MIL The author of the tracker Boris Babenko Doctor presentation MIL How the tracker works under cover .
3. Track retention identification ： The output of object detection is a rectangular array containing objects . however , The object has no attached identifier . for example , In the video below , The detector that detects red dots will output rectangles corresponding to all points it detects in the frame . In the next frame , It will output another rectangular array . In the first frame , A specific point can be determined by the position in the array 10 The rectangle at represents , And in the second frame , It can be in position 17 It's about . When using detection on a frame , We don't know which rectangle corresponds to which object . On the other hand , Tracing provides a way to literally join points ！

#include <opencv2/opencv.hpp>

#include <opencv2/tracking.hpp>

#include <opencv2/core/ocl.hpp>

usingnamespacecv;

usingnamespacestd;

// Convert to string

#define SSTR( x ) static_cast< std::ostringstream & >( \

( std::ostringstream() << std::dec << x ) ).str()

intmain(intargc, char**argv)

{

// List of tracker types in OpenCV 3.4.1

string trackerTypes[8] = {"BOOSTING", "MIL", "KCF", "TLD","MEDIANFLOW", "GOTURN", "MOSSE", "CSRT"};

// vector <string> trackerTypes(types, std::end(types));

// Create a tracker

string trackerType = trackerTypes[2];

Ptr<Tracker> tracker;

#if (CV_MINOR_VERSION < 3)

{

tracker = Tracker::create(trackerType);

}

#else

{

if(trackerType == "BOOSTING")

tracker = TrackerBoosting::create();

if(trackerType == "MIL")

tracker = TrackerMIL::create();

if(trackerType == "KCF")

tracker = TrackerKCF::create();

if(trackerType == "TLD")

tracker = TrackerTLD::create();

if(trackerType == "MEDIANFLOW")

tracker = TrackerMedianFlow::create();

if(trackerType == "GOTURN")

tracker = TrackerGOTURN::create();

if(trackerType == "MOSSE")

tracker = TrackerMOSSE::create();

if(trackerType == "CSRT")

tracker = TrackerCSRT::create();

}

#endif

// Read video

VideoCapture video("videos/chaplin.mp4");

// Exit if video is not opened

if(!video.isOpened())

{

cout << "Could not read video file"<< endl;

return1;

}

// Read first frame

Mat frame;

boolok = video.read(frame);

// Define initial bounding box

Rect2d bbox(287, 23, 86, 320);

// Uncomment the line below to select a different bounding box

// bbox = selectROI(frame, false);

// Display bounding box.

rectangle(frame, bbox, Scalar( 255, 0, 0 ), 2, 1 );

imshow("Tracking", frame);

tracker->init(frame, bbox);

while(video.read(frame))

{

// Start timer

doubletimer = (double)getTickCount();

// Update the tracking result

boolok = tracker->update(frame, bbox);

// Calculate Frames per second (FPS)

floatfps = getTickFrequency() / ((double)getTickCount() - timer);

if(ok)

{

// Tracking success : Draw the tracked object

rectangle(frame, bbox, Scalar( 255, 0, 0 ), 2, 1 );

}

else

{

// Tracking failure detected.

putText(frame, "Tracking failure detected", Point(100,80), FONT_HERSHEY_SIMPLEX, 0.75, Scalar(0,0,255),2);

}

// Display tracker type on frame

putText(frame, trackerType + " Tracker", Point(100,20), FONT_HERSHEY_SIMPLEX, 0.75, Scalar(50,170,50),2);

// Display FPS on frame

putText(frame, "FPS : "+ SSTR(int(fps)), Point(100,50), FONT_HERSHEY_SIMPLEX, 0.75, Scalar(50,170,50), 2);

// Display frame.

imshow("Tracking", frame);

// Exit if ESC pressed.

intk = waitKey(1);

if(k == 27)

{

break;

}

}

}

Publisher ： Full stack programmer stack length , Reprint please indicate the source ：https://javaforall.cn/127650.html Link to the original text ：https://javaforall.cn