MUSCLE-NoE
WP2 Web-Repository

WP2 objectives

The principal objective of this workpackage is to co-ordinate activities that facilitate within and between research workpackage collaboration. This principal objective can be divided into several specific objectives that are placed into 3 groups: evaluation, standards and computation.

• The collection of a large test database of copyright-free multimedia data (video, sound, etc.) and video surveillance data, which will be placed at the disposal of members of the network for use in development and testing.
• The manual annotation of some of the data in the test database (location of scene changes, contents of a scene, behaviour of persons, etc.). This will allow objective evaluation of automatic algorithms.
• The organization of evaluation campaigns through which network members can compare the performance of their systems on the standard test data.

Standards

• Development of an Infrastructure for MultiMedia Metadata Management based on Semantic Web technology.

In order to facilitate interoperability and collaboration among the partners of the NoE, we propose to develop an Infrastructure for MultiMedia Metadata Management based on Semantic Web technology. Through this facility network members should be able to extract MPEG-7 features from their multimedia data (sound, image, video) and store them in an XML database. Furthermore, annotations could be integrated in the system as well as querying of the database. This infrastructure will consist of a number of co-operating units, devoted to feature extraction from multimedia objects, database management, algorithms and annotation handling, and integration. This infrastructure will be designed taking into account the use of the most promising existing tools, open-source software, java-based implementation and multimedia metadata standards. The work within this task includes: Maintenance and updating of a repository on multimedia metadata standards and frameworks Multimedia metadata: investigation of the methods and tools for integrating low-level media specific features to high-level specific semantic terms Prototypisation of a multimedia management infrastructure able to integrate information available within the NoE.

Computation

• To make available software for free use between network members.
• To publish resources on computational methods, such as software repositories and tutorial material.

Demonstrator: 4M - MultiMedia Metadata Management

E-Team on Integration of structural and semantic models for multimedia metadata management

Invited Session to KES2007 Computational Learning Methods for Unsupervised Segmentation (CLeMUS) Vietri sul Mare 12nd-14th Sept. 2007

Web Site Organization

Site tested using last versions of Mozilla, Firefox, Netscape and Internet Explorer browsers able to use XSL-T stylesheets