conference paper

Conference/ProceedingsInternational Conference on Computer Vision Theory and Applications (VISAPP)
Start date17.05.2010
End date21.05.2010
AddressAngers, France
EditorINSTICC Press
Author(s)Tobias Senst, Rubén Heras Evangelio, Volker Eiselein, Michael Pätzold, Thomas Sikora
TitleTOWARDS DETECTING PEOPLE CARRYING OBJECTS: A Periodicity Dependency Pattern Approach
AbstractDetecting people carrying objects is a commonly formulated problem which results can be used as a first step in order to monitor interactions between people and objects in computer vision applications. In this paper we propose a novel method for this task. By using gray-value information instead of the contours obtained by a segmentation process we build up a system that is robust against segmentation errors. Experimental results show the validity of the method.
Key wordsmultimedia application, multimedia analysis, gait analysis, periodicity analysis, people carrying objects, scene interpretation, pattern recognition