How to discover hidden semantics in multimedia content - Gold Sponsor MICO Project

August 12, 2016

With the rapid growth of multimedia content on the Web and in corporate intranets, discovering hidden semantics in raw multimedia content is becoming one of the biggest challenges. At the same time, there is no cost-efficient and simple-to-use product available yet to tackle this challenge. Also the existing multimedia analysis components typically operate in isolation with different standards and data models making it difficult to orchestrate and combine the analysis results of connected and related media objects (text, image, audio, video) in a common standard.

MICO project (short for Media In COntext) aims to solve these problems by providing an integrated platform for cross media analysis, metadata publishing, querying and recommendation. To achieve this, it’s international partners Salzburg Research, Fraunhofer IDMT, InSideOut10, Umea University, University of Oxford, Universität Passau as well as the open source software consultancy Zaizi form an interdisciplinary team involving leading European experts in the fields of Semantic Web and Linked Data technologies, distributed high-dimensional databases, natural language processing, and multimedia extraction.

Hence, MICO combines media analysis technologies with semantic and linked data technologies and standards such as W3C Web-Annotation Data Model and SPARQL to store and query multimedia analysis data in a standardized manner. MICO has also introduced new extensions to semantic web and linked data technologies namely SPARQL-MM which is the multimedia extension for SPARQL and Anno4j which is a Java wrapper API for the web annotation data model.

The project runs from 2013 to 2016 and is partially funded by the European Commission 7th Framework Programme.

The MICO project will be present at the SEMANTiCS conference both with a booth at the marketplace as well as with two separate presentations on the Industry track. Chalitha Perara from Zaizi will do a presentation titled “Cross Media Concept and Entity Driven Search for Enterprise”, while Andrea Volpini from InsideOut10 will talk about “Reimagining news blogging with linked data publishing”.