1 Overall status QMUL: Creation of a web-based application for Image Classification and Relevance Feedback for image retrieval. The system is divided into two main modules: - Client side: The system will be able to handle Image Classification (Module 1) and will incorporate User Relevance Feedback (Module 2) - Server side: To include video shot detection and KF extraction. The target functionalities of the system are listed below: - Text based query input - Initial classification result based on Biologically Inspired classifiers - User Relevance Feedback for improving the image classification - Able to handle video streams, in the sense that if a user selects the most relevant KF, he will have an option to view the video (the relevant shot represented by the KF). - To be able to perform shot detection - To be able to extract Key Frames from the video (which will be used to enable the user play the video) The interface will be a web-page where the user will be able to provide inputs for the user relevance feedback. KU: Work and stabilise SemFs, a virtual filesystem for accessing RDF metadata, that will be further used for semantic based interaction with multimedia. This work has been submitted to the I-Know'06 conference. GU: Development of an interactive video retrieval system. The system allows capturing the interaction data from the users and will make use of it in inferring user intentions. Semantics of interaction is captured and will be used to proactively fetch information for the user. A prototype has been implemented and will be used for experimenting with the TREC VID activity. INA + CWI: Gather real world queries from the customers of INA. We are in the process of translating them in English (from French) and of analysing them in order to extract some general requirements for the semantics annotation of the multimedia material. CWI: Facet browsing has become popular as a user friendly interface to data repositories. In CWI, we have developed our own faceted browser, /facet, that extends the facet browsing paradigm of Semantic Web data in four ways: 1. Users are able to select and navigate through facets of resources of any type and to make selections based on properties of other, semantically related, types. 2. We address a disadvantage of hierarchy-based navigation by adding a keyword search interface that dynamically makes semantically relevant suggestions. 3. The interface of /facet allows the inclusion of facet-specific display options that go beyond the hierarchical navigation that characterizes current facet browsing. 4. The browser works on any RDFS dataset without any additional configuration. These properties make /facet an ideal tool for Semantic Web developers that need a instant interface to their complete dataset. The implementation is based on current Web standards and open source software. A demo is available at: http://e-culture.multimedian.nl/demo/search This work has been submitted to the ISWC 2006 conference, main research track. In order to experiment on different presentation techniques for semantics-based interaction with multimedia, we built on the work we performed on NewsML as part of activity 5.1. We used the conversion tools to convert a dataset of news items (kindly provided by AFP) to their RDF representation. We have then integrated them into the faceted browser, currently under development in CWI. The result can be viewed at: http://media.cwi.nl:8086/facet 2 Issues and Deviations from the Plan None 3 Deliverables and Milestones for the Reporting Period None 4 Dissemination The work done on the facetted browser by CWI has been submitted to the ISWC 2006 conference, main research track. 5 Meetings Attendance KU has presented a poster of the SemFs system during the 3rd European Semantic Web Conference (ESWC'06), http://www.eswc2006.org/ (11-14 of June) CWI has presented the results of the NewsML work integrated in the Facetted Browser during the IPTC Annual General Meeting, in Vienna, http://www.iptc.org/pages/meetg_main.php (2-6 of July). 6 Conclusion Everything is going as planned