Personalized Content Presentation for Virtual Gallery

Abstract

Utilizing Virtual Reality technologies for virtual museum brings new ways of interactive presentation of the contents. In addition to interactivity, personalization is an important emerging issue in digital content management especially with virtual reality. For the virtual museum or gallery, selection and presentation of personalized content will improve user experience in navigating through huge collections like Musee du Louvre or British Museum. In this paper, we present a personalization method of massive multimedia content in virtual galleries. The proposed method is targeted for the pictures that could be characterized by its large amount of source in galleries. The method is based on classified image features which are extracted using standard MPEG-7 visual descriptors. Using Neural Networks, the best matching pictures are selected and presented in the virtual gallery by choosing similar styles from the extracted preference features. The simulation results show that the proposed system successfully classifies images into correct classes with the rate of over 75% depending on the employed features. We employ the result into a virtual gallery application which gives opportunities of automatically personalized gallery browsing.