A Ball Tracking Application

The tracker implements the Covariance Tracking method to track the ball in a football match

Description

The application uses the approach introduced in paper "Covariance Tracking using Model Update Based on Means on Riemannian Manifolds", F.Porikli, O.Tuzel, P.Meer.

The tracking is based on:
1) initializing the target region
2) constructing the Feature Vectors for each pixel in the target region (first frame)
3) forming the Covariance Matrix using feature vectors generated in step 1
4) determining the candidate regions for the following frames and constructing the covariance matrices of these regions
5) finding the minimum covariance-distanced region from these candidate region
6) assign this region as the estimated region

• MATLAB release: MATLAB 7.6 (R2008a)

[Ronak]

hey i tried your code. But it showed an error,
??? Error using ==> aviread at 76
Unable to locate decompressor to decompress video stream
for the command:
mov = aviread(videoFile, 1:maxFrame);
but i got a solution for it, using mmreader....

readerobj = mmreader('tv cam-9.avi');
vidFrames = read(readerobj);
for k = 1 : maxFrame
    mov(k).cdata = vidFrames(:,:,:,k);
    mov(k).colormap = [];
end

[Ronak]

With the above comment I meant that the error was coming in the trackball.m file, with line47:

mov = aviread(videoFile, 1:maxFrame);

And I solved it by replacing it with following lines:

readerobj = mmreader('tv cam-9.avi');
vidFrames = read(readerobj);
for k = 1 : maxFrame
    mov(k).cdata = vidFrames(:,:,:,k);
    mov(k).colormap = [];
end

And the modified code works just fine.

[Sisir Kanti]

Not working! Same errors.

@Ronak:
Your functin vidFrames not found!

Some can help please reply here!

[praneeth botta]

can you please help,
i cannot understand CovarianceMatrix , eig and temp ...
can someone help me understanding ..

[Badri Subudhi]

i am also getting same problem

[A. Bhargav Anand]

Guys It works for me, but the thing is it displays all frames as images.
What if we want to track the ball in real time ????

[SUDHAKAR B]

 

[abhijith bailur]

how to run this file i am finding diffcult please help me out

[Michael]

I have used findCovarianceMatrix.m,
and it works fine, although it is not efficiently implemented e.g. it is not using the integral image method. However I do not see any function to deal with the computation of a minimum covariance-distance.

[bhargav m]

I have a problem in understanding the initial position of the object, How is it calculated?
what does position offset mean?

[Ehsan]

Sorry, the last bit in my previous comment (I had only one question!) was added by mistake!

[Ehsan]

?

[Ehsan]

For some reason, my comment was deleted! don't know why, sorry!

this was my comment which is apparently deleted!

Thanks Fuat for the code, very helpful! I just wanted to ask if you implemented the update mechanism which is based on Riemannian manifold. I guess it's not included in the code! Am i right?