K-Space Content Management and Retrieval System

In this contribution, the so-called "k-space content management and retrieval system " is presented. This multimodal framework is being developed within the K-Space project. K-Space is a network of leading European research teams from academia and industry conducting integrative research and dissemination activities in semantic inference for automatic and semi-automatic annotation and retrieval of multimedia content. To the large number of the system functionalities count: automatic shot detection, key-frame extraction, mpeg-7 feature extraction, multimedia content annotation, video streaming, visual search and retrieval, relevance feedback, and visual content classification. First, the system features from the user point of view are presented. Second, the theoretical description of algorithms leading to these functionalities follows.