Find the most recent information on EU Funding activities in the field of Information and Communication Technologies (ICT) by visiting our ICT in FP7 website, which covers ICT in the 7th Framework Programme (FP7) 2007 - 2013.
SEMEDIA - Search Environments for Media
SEMEDIA started in January 2007, and the overall objective is to develop a collection of audiovisual search tools that are heavily user driven, preserve metadata along the chain, and are generic enough to be applicable to different fields (broadcasting production, cinema postproduction, and social web).
Impact
The overall objective of SEMEDIA is to create new methods, environments and widely usable tools for media labelling, searching and retrieval from very large collections of heterogeneous data, building on and extending research in media technologies, web semantics, AI, content-based IR and interface design.
Main innovation
The results will be expressed as innovations in the form of:
- retrieval and data-mining models for multimedia objects;
- plug and play modular search tools to sit on a wide range of media management platforms;
- query interfaces that integrate secure access and rights management; and
- usage based annotation and feedback models for multimedia objects.
SEMEDIA aims to develop a collection of audiovisual search tools that are heavily user driven, preserve metadata along the chain, and are generic enough to be applicable to different fields (broadcasting production, cinema postproduction, and social web).
Results so far
The project is in its starting phase. We are now conducting interviews and collecting questionnaire data in order to develop initial scenarios and user requirements for search and retrieval methods in the different contexts of broadcast television, postproduction (across film, advertising and new media), web-based social exchange and a future merged broadband environment that draws on TV archives and public sources of data.
More details
- Prototype Poster (ZIP, 2.882KB)
- Annual Report 2008 (PDF, 2.882KB)
- Annual Report 2007 (PDF, 342KB)
- SEMEDIA project website
Upcoming work
SEMEDIA is planned as a thirty-month work programme, structured as seven work packages. The specialised Research is grouped into two large work packages, WP3 Media Content Analysis and Representation and WP4 User Behaviour and Interaction in Media Search. The focus of these work-packages and their constituent tasks represents the key areas where the project expects to make fundamental advances. The research is based on and directed by WP2 Scenarios of Use, which will carry out detailed task analyses and define requirements driven by the media industry partners, in a wide range of industry contexts with a vision for the future. The development of an integrated architecture, practical tools and technologies, which incorporate the research results and can be tested against the Scenarios, is an iterative activity and will employ rapid prototyping methods conducted in WP5 Prototypes of Media Search Environments. The prototypes produced will be evaluated in laboratory benchmark tests and by in-depth usability tests by our User partners (CCRTV, BBC, and S&M) and by the web based community (Y!I).
In summary, the SEMEDIA project workflow addresses the following principal elements: < ul>
- User requirements analysis based on initial use scenarios to confirm the functions, features and performance required by creative personnel in a variety of application scenarios covering media search and retrieval tasks (including their workflow, security, interoperability and interface requirements) in broadcast television; film, TV and advertising postproduction; media archives, and web-based communities.
- Development and refinement of the functional scenarios, based on observation and trials of early prototypes with real media content, processes and workflows.
- Researching appropriate methods and procedures for analysing, formatting and representing searchable media, including the summarising of sparsely annotated content, low-level indexing for more efficient searching, metadata and relevance feedback features, data architectures and security issues, and media mining techniques.
- Researching new methods and interfaces for interaction based content annotation and metadata extraction, feedback models, context-aware and intent-aware adaptive search, feedback-based search, and models for integrating retrieval and mining methods.
- Design and development of software prototypes, based on open architectures, and their integration as functional prototypes of search environments with a wide range of existing media production and management environments to prove their effectiveness.
- Evaluating prototype search and retrieval environments across the range of professional film, television, advertising, archive and web media contexts.
- Disseminating results progressively throughout the media industries and planning the exploitation of new knowledge, products, open systems and methods for media search environments.
Administrative Details
- SEMEDIA (IST-045032) is a Specific Targeted Research Project of the European Union's 6th Framework Programme - call 6.
- The project started on 1 January 2007 and finishes on 30 June 2009.
- There are 8 partners from 4 European countries involved in the project, and the overall budget is 2.73 million euro.
List of Participants
- Project Coordinator: Fundació Barcelona Media Universitat Pompeu Fabra (FBM-UPF), SPAIN
- CCRTV Administració Sistemes d'Informació, S.A (CCRTV-ASI), SPAIN
- JOANNEUM RESEARCH Forschungsgesellschaft mbH (JRS), AUSTRIA
- The University of Glasgow (UG), UK
- DVS Digital Video Systems GMBH (DVS), GERMANY
- Smoke and Mirrors Ltd (S&M), UK
- Yahoo! Iberia (Y!I), SPAIN
- British Broadcasting Corporation (BBC), UK
- Universitat Politecnica de Catalunya (UPC), Spain
Contact Person
- Project coordinator:
- Ricardo Baeza-Yates
- Fundació Barcelona Media Universitat Pompeu Fabra
Events in connection with SEMEDIA
- New Challenges in Audio Visual Search. Multimedia Information Retrieval, SIGIR Workshop, Amsterdam, NL, 27 July 2007.
- SSMS07. Summer School on Multimedia Semantics, Glasgow, UK, 15-21 July 2007.
- SAMT 2007. Semantic and Digital Media Technologies, Genova, Italy, 5-7 December 2007.