webcam - How to track the area within a rubberband rectangle in a video as it streams using C# -

I am doing a project for User Initial Real Time Object Tracking in C #. What do I want to do, take a photo from the webcam into a picture box (done by Dishnet), then draw a rubber band rectangle on the video (say a person's face / eye / nose / whole body - I think I I'm going to radiate the mouse), then I want to track the area attached to the rubber band rectangle.

I'm currently going through a sample dshownet bitmapmixer sample in the sample folder to learn to draw on a video (though I have made a rubberband rectangle, though it is no success yet, it is now the top of the video It's a matter of working).

My main point is to track, which is attached within the rubber band rectangle (after rectangular drawn it As long as a button command is not erased). Someone has said (I really do not know in my project) to go through OK, I can be wrong

The way I think about it, that's it, I think if I can see the area outside the rectangle as a background, and the color of the area can be removed in the histogram within the rectangular (foreground) and check whether it goes fashion in later frames, I Tapurwk could track me [I would have achieved exactly how to do it through do not know the code it] Is it true ??

By the way, I consider the tracker to be rectangular, which is streaming video until it becomes visible in the video and proceed with the person. To get started, I am using this with a saved video file.

How to track any ideas? Does it matter no matter what type of face or body I am a part of human being to track its movements?

Thank you for your time.

It looks like what you need to find a human face in each frame and a rectangle Cover area I am not an expert in face detection, but I am using auto satire production project. Analyzing the color histogram is a way to do this, but I remember that the shape-based approach was working better. Also, check it because it is a better model. If you google them, you should be able to find many articles. BTW, if you need a quick way, I would advise to use it. To find many faces in an image and get those areas as rectangles, there is already very good and simple task.