Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Video segmentation via a Gaussian switch background model and higher order Markov Random fields

 
: Radolko, Martin; Gutzeit, Enrico

:
Volltext urn:nbn:de:0011-n-3367263 (729 KByte PDF)
MD5 Fingerprint: 2241686e80533c18fff88e5d737f71cc


Braz, José (Ed.); Battiato, Sebastiano (Ed.); Imai, Francisco (Ed.) ; Institute for Systems and Technologies of Information, Control and Communication -INSTICC-, Setubal:
10th International Conference on Computer Vision Theory and Applications, VISAPP 2015. Proceedings. Vol.I : Berlin, Germany, 11 - 14 March 2015; Part of VISIGRAPP, the 10th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
SciTePress, 2015
ISBN: 978-989-758-089-5
pp.537-544
International Conference on Computer Vision Theory and Applications (VISAPP) <10, 2015, Berlin>
International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP) <10, 2015, Berlin>
English
Conference Paper, Electronic Publication
Fraunhofer IGD ()
image segmentation; Markov random fields (MRF); video segmentation

Abstract
Foreground-background segmentation in videos is an important low-level task needed for many different applications in computer vision. Therefore, a great variety of different algorithms have been proposed to deal with this problem, however none can deliver satisfactory results in all circumstances. Our approach combines an efficient novel Background Substraction algorithm with a higher order Markov Random Field (MRF) which can model the spatial relations between the pixels of an image far better than a simple pairwise MRF used in most of the state of the art methods. Afterwards, a runtime optimized Belief Propagation algorithm is used to compute an enhanced segmentation based on this model. Lastly, a local between Class Variance method is combined with this to enrich the data from the Background Substraction. To evaluate the results the difficult Wallflower data set is used.

: http://publica.fraunhofer.de/documents/N-336726.html