- research themes
- research groups
- Algorithms and Complexity
- Computational Dynamics
- Cryptology
- Database Architectures
- Distributed and Interactive Systems
- Formal Methods
- Information Access
- Intelligent Systems
- Life Sciences
- Multiscale Dynamics
- Networks and Optimization
- Scientific Computing
- Software Analysis and Transformation
- Stochastics
- research staff
- In brief
K-Space
Project name: EU IST Network of Excellence: K-Space - The Knowledge Space of Technology to Bridge the Semantic Gap
Research group: Interactive Information Access (INS2)
Coordinator of this project: Raphael Troncy
The Semantic Gap is a term used to denote the gap between feature descriptors that can be computed automatically by current machines and algorithms, and the richness and subjectivity of semantics in high-level human interpretations of audiovisual media. The EU K-Space project tackles this issue by integrating leading European research teams to create a Network of Excellence in semantic inference for semi-automatic annotation and retrieval of multimedia content.
Semantic Media Interfaces will lead the research in the area of "Semantics-based Interaction with Multimedia", including the automatic generation of human readable presentation of multimedia material and the retrieval of the relevant data by querying, filtering and processing the associated semantic annotation. In addition, we will participate in the development of a multimedia ontology infrastructure, in the research for representation of content and context of multimedia data, as well as the storage of content and context in an appropriate metadata repository.
The joint research activities of the network are aimed at convergence and resources optimization by exploiting multidisciplinary aspects of multimedia knowledge extraction. The research results of project partners will be integrated to deliver a framework for collaborative research on knowledge acquisition and modelling for multimedia content. The research is divided into three main areas, to be combined at a later stage.
- Content-based multimedia analysis: Tools and methodologies will be developed for low-level signal processing, object segmentation, audio processing, text analysis, and audiovisual content structuring and description. We have developped the VAMP service (http://vamp.joanneum.at/) that can semantically validate MPEG-7 descriptions.
- Knowledge extraction: A multimedia ontology infrastructure will be constructed to exploit the potential for knowledge acquisition from multimedia content, knowledge-assisted multimedia analysis, context based multimedia mining and intelligent exploitation of user relevance feedback. COMM, a Core Ontology for MultiMedia, is available at: http://multimedia.semanticweb.org/COMM/.
- Semantic multimedia: Knowledge representation for multimedia, distributed semantic management of multimedia data, semantics-based interaction with multimedia and multimodal media analysis will be investigated and developed. As an example, we are working on converting NewsML G2 into semantic web technologies in order to better present news stories, pictures and videos to end users. See a demo at: http://newsml.cwi.nl/
The work carried out within the boundaries of the project is also closely related to the Multimedia Annotation on the Semantic Web Task Force of the W3C Semantic Web Best Practices and Deployment Working Group, and the W3C Multimedia Semantics Incubator Group.
Members
Alia Amin, Lynda Hardman, Michiel Hildebrand, Zeljko Obrenovic, Jacco van Ossenbruggen, Raphael Troncy.
Key publications of K-Space
- Richard Arndt, Raphaël Troncy, Steffen Staab, and Lynda Hardman. Adding Formal Semantics to MPEG7: Designing a Well-Founded Multimedia Ontology for the Web. (KU and CWI technical report KU-N0407), January 2007.
- Zeljko Obrenovic, Raphaël Troncy, and Lynda Hardman. Vocabularies for Description of Accessibility Issues in Multimodal User Interfaces. In: Ielka van der Sluis et al. (eds.): MOG 2007 - CTIT Proceedings of the Workshop on Multimodal Output Generation (pages 117-128), January, 25-26, 2007, Aberdeen, UK.
- Raphaël Troncy, Werner Bailer, Michael Hausenblas, Philip Hofmair, and Rudolf Schlatte. Enabling Multimedia Metadata Interoperability by Defining Formal Semantics of MPEG-7 Profiles. In: Semantic Multimedia - SAMT 2006 (pages 41-55), December 6-8, 2006, Athens, Greece.
- Lynda Hardman and Jacco van Ossenbruggen. Creating Meaningful Multimedia Presentations. In: IEEE International Symposium on Circuits and Systems, May 2006, Special session on Advances in semantic multimedia analysis for personalised content access
Reports
CWI reports of K-Space
Browse
http://www.k-space.eu/
