History The advent of this action is not at all surprising considering both the strategically important and timely nature of the technical objectives addressed and the close relationships and long history of collaboration of its participants. The participants of this proposal have collaborated in a number of initiatives in the past, most notably the COST 211 action that targeted Redundancy Reduction Techniques and Content Analysis for Multimedia Services. The COST 211 action was initiated in the eighties and it's original research focus was on normative aspects of video compression. It contributed directly to the development of ITU-T compression standards such as H.261 and H.263 via the collaboratively developed COST software Reference Model (RM) as well as individual partners' contributions to ITU-T. Whilst the focus of COST 211 was on normative aspects, the action also acted as a forum for information exchange and discussion on non-normative aspects of compression and related activities. As such, it was not surprising that with the advent of the more recent work items of the ISO/IEC Moving Picture Experts Groups (MPEG) in subsequent years, COST 211 was able to seamlessly adapt its research. COST 211 collaboratively developed a software Analysis Model (AM) for segmentation and tracking and made this available to the wider research community in the context of a segmentation evaluation framework. COST 211 ended in May 2003. The members of this proposal were all members of COST 211 at various stages of its evolution. Over the years, COST 211 acted either as a spring-board for other collaborative initiatives or as forum for such initiatives to come together. Examples are the RACE MAVT (Mobile Audio Visual Terminal) and MORPHECO (Morphological Codec for Storage and Transmission) projects and the ACTS MoMuSys (Mobile Multimedia System) project. Other more recent examples include the ACTS DICEMAN (Distributed Internet Content Exchange with MPEG-7 and Agent Negotiations), SAMBIST, BUSMAN (Bringing User Satisfaction to Media Access Networks), etc. Again, many partners in the team that prepared this proposal worked on these or similar projects such as MUVIS framework. In 1998, first MUVIS system has been implemented for indexing large image databases and retrieval via search and query techniques based on semantic and visual features. Based upon the experience and feedback from this first system, recently a new PC-based MUVIS system, which is further capable of content-based indexing and retrieval of video and audio information in addition to several image types have been developed. Clearly, COST 211 was very successful in mobilising research activities within Europe in a flexible and reactive manner. However, given the shift in focus in European research in the area of digital content analysis, it is clear that COST 211 is no longer the best instrument for continuing this success. A completely new action is now required. A project with the objective of developing new technology for bridging the semantic gap and knowledge extraction from audiovisual content such as this, requires a new blend of partners and expertise as yet unavailable elsewhere. Given the partners' reputations and track records in the academic community and leveraging the fact that they have worked together on proposals in related areas, it was relatively straightforward to produce the objectives and technical work-plan outlined later in this document. The proposal document itself was collaboratively developed by all partners over email. The fact that many partners attend the same project meetings and/or conferences was leveraged whenever possible in order to discuss and progress the project proposal. The proposal structure (along with responsible assigned to each section) was initially formulated by the co-ordinators and then distributed to all partners. Contributions were then collected and edited in order to produce the final document.