Plenary Lecture, ARTIFICIAL INTELLIGENCE, KNOWLEDGE ENGINEERING and DATA BASES (AIKED '09), Cambridge, UK, February 21-23, 2009

Plenary Lecture

Semantic Approaches for Web Multimedia
Processing and Communication




Assistant Professor Ashraf M. A. Ahmad
Department of Computer Graphics and Courtesy Department of Computer Science
Princess Sumya University for Technology
P. O. Box 1438 Amman 11941, Jordan
E-mails: ashrarf@csie.nctu.edu.tw, a.ahmad@psut.edu.jo


Abstract: The motivation for emerging the concept semantic web is due to the fact that search engines and text-based exploring are no longer adequate as these approaches involve an extensive information retrieval process. The deployed searching and retrieving descriptors are naturally subjective and their deployment is often restricted to the specific application domain for which the descriptors were configured. The new era of information technology imposes different kinds of requirements and challenges. Automatic extracted audiovisual features, as these features are more objective, domain-independent and more native to the audiovisual content are required. Therefore, this presentation is a detailed trail to introduce to the researchers, experts, students and practitioners very valuable references and guides to lead them through their exploration and researching the multimedia content and semantic web. The results of leading researchers from around the world are successfully collected in one concise presentation in very semantic manner.
This highly informative presentation includes three parts to construct a valuable reference resource for researchers, student and experts dealing with the emergence of integrated semantic multimedia field.
The presentation introduces the concept of semantic web and multimedia content analysis to the readers through a logical sequence from standards and hypothesis through system examples, presenting relevant tools and methods. In addition, this presentation addresses approaches applied to specific problem domains such as content search, indexing and retrieval.
The attendees may attain sufficient knowledge to start projects or researches related to the presentation theme from finding recent results and articles related to the active research area of integrating multimedia with semantic web technologies.
The semantic web technologies are elaborated and ontology representation has been emphasized. The presentation provides an excellent summary of the fundamental theory behind applying a knowledge engineering approach to vision problems. This summery iconize the concept of semantic web and multimedia content analysis. A definition of the fuzzy knowledge representation that can be used for realizing on multimedia content applications has been provided with comprehensive analysis.
The second part of presentation introduces the multimedia content analysis approaches and applications. In addition, some examples of methods applicable to multimedia content analysis are presented.
Multimedia content analysis is very diverse and concerns many other research fields at the same time. Multimedia content analysis has strong diversity issues, as it starts from low level features such as (colors, DCT coefficients, motion vectors, etc.) up to the very high and semantic level (Object, Events, Tracks, etc.). The second part includes topics on structure identification (e.g. shot detection for video sequences), and object-based video indexing. These conventional analysis methods are supplemented by results on semantic multimedia analysis, including development and use of knowledge models for automatic multimedia analysis.
The final part of the presentation discusses and describes example systems which current projects have been implemented, and include extensive results and real demonstrations. For example, real case scenarios like ECommerce medical applications and web services have been introduced. Natural language, speech and Image processing techniques and its application for multimedia indexing and contentbased retrieval have been elaborated with extensive examples and deployment methods.

Brief Biography of the Speaker:
Ashraf Ahmad obtained his PhD degree in Computer Science and Engineering from National Chiao Tung University (NCTU) in Taiwan. He obtained his B.Sc. degree from Princess Sumya University for Technology (PSUT) in Jordan. Dr. Ahmad is currently an assistant professor at the department of computer Graphics in PSUT, Jordan. His interest area includes multimedia semantic features extraction, and analysis, multimedia retrieval, and multimedia communication. Prof. Ahmad has authored over 50 scientific publications including journal papers, conference papers and book chapters. In addition, Dr. Ahmad has several US and international patents in his field of expertise. He serves in program committee for several international conferences. He is also a reviewer and referee for several conferences and journals. His work has been published and presented at various international conferences.
Dr. Ahmad has been listed in Who's Who in the World for the year 2006 and Who's Who in Asia for the year 2007. In addition, He has been elected as one of the 2000 Outstanding Intellectuals of the 21st Century for the year 2006 for his outstanding contribution in field of Video Processing and Communications. Ashraf Ahmad has been chosen as one of the recipients of Leading Scientist award in the year 2006. In 2008 Dr. Ahmad and his team won the first place in both Jordan and Middle East Level in the world wide international competition Imagine Cup for their novel project in Air Pollution Detection using video analysis. Again Dr. Ahmad and his team won the first place in METS’08 competition for their project in Bluetooth Mobile Application field.
 

 

 

WSEAS Unifying the Science