Results of the Industry-Research co-operation on Multimedia Analysis and Annotation
Chosen Use Case: Automated Semantic Multimedia Annotation
KW Partners: CERTH
IB Member: Motorola
Progress
Semantic Web technologies have been successfully employed for the representation of both media-related and domain-specific knowledge, including contextual
knowledge, in a multimedia analysis and annotation system. This work is part of the EU project aceMedia, see
http://www.acemedia.org
The aceMedia use case addresses semantic analysis, annotation and retrieval of multimedia content. It uses ontologies for the representation of domain
specific and media related information, contextual knowledge representation, logic-based reasoning for the integration of content annotations and the
extraction of implicit knowledge, and a customized approach for representing and reasoning under uncertainty.
Evaluation
The aceMedia use case encompasses a variety of industry requirements with respect to multimedia content annotation, analysis and management.
The implemented algorithms and demos have demonstrated the applicability of Semantic Web technologies towards a uniform representation formalism
that facilitates modules’ interaction and information exchange, the potential of fuzzy extensions to partially meeting the ambiguity in multimedia,
and a formal representation of contextual knowledge.
Results
The aceMedia project constitutes a pragmatic demonstration of Semantic Web technologies application in multimedia content analysis and annotation.
Although not providing a complete solution, the implemented demo versions show the value and further potential. At the same time, the experiences gained,
form valuable lessons for future directions.
In order to be industry ready, it requires a complete, robust solution, whose formalization would ensure the generality of application and its
extensibility. As an important part of the research relates to the expressiveness of current Semantic Web languages and corresponding tools support,
porting the aceMedia approach directly to industry strongly interrelates with the respective Semantic Web advances.
Available results include:
References
G. Stoilos, G. Stamou, V. Tzouvaras, J.Z. Pan, I. Horrocks, A Fuzzy Description Logic for Multimedia Knowledge Representation, Proc. of the International
Workshop on Multimedia and the Semantic Web.
K. Petridis, S. Bloehdorn, C. Saathoff, N. Simou, S. Dasiopoulou, V. Tzouvaras, S. Handschuh, Y. Avrithis, I. Kompatsiaris and S. Staab: Knowledge
Representation and Semantic Annotation of Multimedia Content, IEEE Proceedings on Vision Image and Signal Processing, Special issue on Knowledge-Based
Digital Media Processing, Vol. 153, No. 3, pp. 255-262, June 2006.
S.Dasiopoulou, C. Saathoof, Ph. Mylonas, Y. Avrithis, Y. Kompatsiaris, and S. Staab, Introducing Context and Reasoning in Visual Content Analysis: An
Ontology-based Framework, in Semantic Multimedia and Ontologies: Theories and Applications, Paola Hobson, Ioannis Kompatsiaris (Editors), Springer-Verlag,
2007, to appear.
For more information about this Industry-Research co-operation please contact
Dr Lyndon Nixon, Knowledge Web Industry Area co-manager nixon@inf.fu-berlin.de