The Application of Artificial Intelligence Techniques in Multimedia Databases
The Application of Artificial Intelligence Techniques in Multimedia Databases
Payam Mamaani Barnaghi and Sameem Abdul Kareem
Department of Artificial Intelligence
Faculty of Computer Science and Information Technology
University of Malaya
Kuala Lumpur
Abstract:
Multimedia data are pre-eminent and prominent parts of information systems nowadays. The extensive progress in hardware and software technologies has led to the storage and management of huge amounts of multimedia data in database systems. Although multimedia data is a widely used aspect of information systems, multimedia database applications rely on traditional approaches to retrieve the data. The keyword based information discovery methods provide search and retrieval facilities for multimedia objects, but, there is a wealth of contextual and semantic information hidden in the content of multimedia objects which may not be addressed by keyword searches. Artificial Intelligence (AI) techniques provide a means to extract and analyze content-based information and retrieve the implicit data in multimedia objects. The feature detection algorithms for content-based retrieval (based on color, texture for image data and etc.), structure analysis (based on shapes and spatial features and etc.), and automatic annotation of semantics in multimedia data are examples of using AI techniques in multimedia database applications. The application of AI techniques in multimedia databases allows database developers to investigate the relationships amongst multimedia objects and to provide a way for representing the semantic information in these databases. This content based representation provides effective and easier interpretation as well as search and retrieval mechanisms for multimedia database contents. An intelligent presentation generation is also another emergence of AI techniques in multimedia database applications. Intelligent presentation process generates the presentation structure based on the semantics and express how the objects will interact with each other to provide understandable and efficient presentations. This paper surveys AI techniques, multimedia database systems, multimedia data modelling; content-based multimedia representation and presentation generation systems. The primary focus is on content-based multimedia representation projects and the efforts in providing intelligence in hypermedia presentation generations.
