SOMIR
Semantic Oriented Multimedia Indexation and Retrieval
| Coordinatore | UNIVERSITE PAUL SABATIER TOULOUSE III
Organization address
address: ROUTE DE NARBONNE 118 contact info |
| Nazionalità Coordinatore | France [FR] |
| Totale costo | 0 € |
| EC contributo | 220˙594 € |
| Programma | FP7-PEOPLE
Specific programme "People" implementing the Seventh Framework Programme of the European Community for research, technological development and demonstration activities (2007 to 2013) |
| Code Call | FP7-PEOPLE-IEF-2008 |
| Funding Scheme | MC-IEF |
| Anno di inizio | 2009 |
| Periodo (anno-mese-giorno) | 2009-05-01 - 2011-04-30 |
Partecipanti
| # | ||||
|---|---|---|---|---|
| 1 |
UNIVERSITE PAUL SABATIER TOULOUSE III
Organization address
address: ROUTE DE NARBONNE 118 contact info |
FR (TOULOUSE CEDEX 9) | coordinator | 220˙594.23 |
Mappa
Word cloud
Esplora la "nuvola delle parole (Word Cloud) per avere un'idea di massima del progetto.
Obiettivo del progetto (Objective)
'The main purpose of the present project is to provide a solution for the efficient semantic-oriented multimedia information retrieval with the support of both the traditional multimedia indexing and classification techniques (based on processing the multimedia content itself) and the modern semantic Web technologies (oriented to exploring the associated multimedia metadata). The challenge proposed by these research objectives concerns a solution for the problem of bridging the “semantic gap” between low-level multimedia features and high-level concepts describing the content. The various metadata types situated in this interval will be organized into a three layers structured collection: generic metadata, valid for all multimedia types; media-specific metadata; semantic metadata for describing the multimedia objects content with the support of various specific vocabularies and ontologies. Supplementary semantic information will be managed through a knowledge database. Aiming to integrate the existing heterogeneous set of automatic indexation and classification algorithms, a structured algorithms collection will be defined and developed. A generic interface for the algorithms will be adopted, where their input/output data, as well as their preconditions and effects will be correlated with the structured metadata collection. Alongside to searching into metadata collection, three issues will be considered when providing results to the user complex queries: to process the query, to define the algorithms sequences in order to be successively applied for indexing the multimedia content, and to use the information provided by the knowledge database. For gaining an user-centered dimension, our approach will define and develop an user profile expressing his long-term interests and his current activity goals in terms of multimedia structured metadata. The profile will be considered for refining the user query’s results and to provide him with personalized recommendations.'
Introduzione (Teaser)
New online search technology and algorithms are set to help users more easily find what they want in the growing sea of online multimedia content.
Descrizione progetto (Article)
The Internet has become a huge repository for multimedia information, i.e. a combination of audio, video and photos, from scientific presentations to music video clips. As this online treasure grows, it is important to develop easier and more efficient ways to search for relevant multimedia content.
The EU-funded project 'Semantic oriented multimedia indexation and retrieval' (SOMIR) sought to address the challenge of obtaining desired multimedia information. It employed a semantic-oriented approach that considered a combination of multimedia metadata, indexing algorithms and user queries to achieve accurate searches.
In more detail, the project investigated how a user's query is executed to develop more efficient content indexation, which is crucial for managing and retrieving relevant information. It looked at indexing based on algorithms that generate diverse, heterogeneous multimedia metadata. The latter could refer to language the content is written in, tools used to create it and links on the subject, enabling browsers to improve the experience of users.
SOMIR improved the relevance of query results through query enhancement and/or query rewriting, as well as by indentifying the relevant algorithms to improve indexation and querying. At the end, the technology used query histories to create personalised search recommendations for users.
The results of the project were published in numerous articles, notable journals and papers. They may ultimately help third parties upgrade the way multimedia content reaches its audience and enhance the information age in many ways.