Management of Images Collections

The proliferation of handled devices combined to telecommunication and information technology allows to manage visual documents from everywhere at any time. The management methodologies need to take into account different factors and constraints such as battery life, processing capacity, communication costs, user specificity, and user context. In our talk we will focus on content based visual document suggestion (CBIS). An unified statistical model for the user need, his/her context, and the visual document collection will be presented. The suggestion is based on Bayesian reasoning such as the estimation of mixtures of Dirichlet pdfs. A mixture of Dirichlet pdfs have nice properties for modeling a non-Gaussian data in high dimensional space. Several experimental results will be shown.

Dr. Djemel Ziou

Professor of Computer Science, Universite of Sherbrooke, QC, Canada on September 21, 2007 at 12:30 PM in Engineering Building II, Room 1230

Dijemel Ziou received the B.Eng. degree in Computer Science from the University of Annaba in 1984, and Ph.D. degree in Computer Science from the Institut National Polytechnique de Lorraine (INPL), France in 1991. Presently, he is full Professor at the Department o Computer Science, Universite de Sherbrooke, QC, Canada. He is holder of the NSERC/Bell Canada Research Chair in personal imaging. His research interests include image processing, information retrieval, computer vision and pattern recognition.

The Department of Electrical and Computer Engineering hosts a regularly scheduled seminar series with preeminent and leading reseachers in the US and the world, to help promote North Carolina as a center of innovation and knowledge and to ensure safeguarding its place of leading research.