Background Modelling
I am working at a project in my college.
I have gone through enough progress in the filed of background modelling and I am aware of the state of the art in this field.
I have figured out a proper way to do it, which will be along the lines of maintaining gaussians.
I wanted to know if there is any fine difference between the appraoches if i want to model the backgroud or i want to extract foreground to do some learning on it.
thanks in advance.
- Login or register to post comments
- Printer-friendly
Dear Vishal,
The two are related as Jai said. You have to chose the background model following the critical situations presented in the video such as waving trees, illumination changes. Generally, the foreground detection depend on the background model.
See the surveys:
T. Bouwmans, F. El Baf, B. Vachon, “Background Modeling using Mixture of Gaussians for Foreground Detection - A Survey”, Recent Patents on Computer Science, Volume 1, No 3, pages 219-237, November 2008.
T. Bouwmans, F. El Baf, B. Vachon, “Statistical Background Modeling for Foreground Detection: A Survey”, Handbook of Pattern Recognition and Computer Vision, World Scientific Publishing, Volume 4, Part 2, Chapter 3, pages 181-199, January 2010.
and more information here:
http://sites.google.com/site/backgroundsubtraction/Home
Best regards!
Thierry
Hello Vishal
The two are related. If you can model the background well, it will help you in foreground object extraction. A simple way is to look for regions not following the model.
If you want to extract the foreground and if it has some consistent properties, like that of a human face, you can model that as well, which will improve your results.
Good Luck
Jai
Graduate Research Assistant
University of Maryland, College Park
http://www.umiacs.umd.edu/~jsp