Objective CLASS is a basic research project focused on developing a specific cognitive ability for use in intelligent content analysis: the automatic discovery of content categories and attributes from unstructured content streams. More specifically, it will study object recognition and scene analysis in images, video, and accompanying text streams. Current visual recognition methods either work at unsatisfyingly low semantic levels, or require large amounts of manually labelled data for training. Autonomous learning will make them more adaptive and allow more general classes and much larger and more varied data sets to be handled. This will be useful in many applications of cognitive systems and intelligent agents that handle streams of sensed data.We will study both fully autonomous and semi-supervised methods at three levels of abstraction: new individuals (specific people, objects, scenes, actions); new object classes and attributes; and hierarchical categories and relations between entities. The work will combine robust computer vision based image descriptors, machine learning based latent structure models, and advanced textual summarization techniques. The potential applications of the basic research results will be illustrated by three proof-of-concept demonstrators: an Image Interrogator that interactively answers simple user-defined queries about image content; a Video Commentator that automatically creates textual descriptions of the action and content of situation comedy videos for visually impaired users; and a News Digester that combines television news stories with captions from several sources to create a textual and visual digest of them.The CLASS consortium is interdisciplinary, combining five leading European research teams in visual recognition, text understanding and summarization, and machine learning. Fields of science natural sciencescomputer and information sciencesartificial intelligencecomputer visionnatural sciencescomputer and information sciencesartificial intelligencemachine learning Programme(s) FP6-IST - Information Society Technologies: thematic priority under the specific programme "Integrating and strengthening the European research area" (2002-2006). Topic(s) IST-2004-2.4.8 - Cognitive Systems Call for proposal Data not available Funding Scheme STREP - Specific Targeted Research Project Coordinator INSTITUT NATIONAL DE RECHERCHE EN INFORMATIQUE ET EN AUTOMATIQUE EU contribution No data Address DOMAINE DE VOLUCEAU 78153 LE CHESNAY France See on map Total cost No data Participants (4) Sort alphabetically Sort by EU Contribution Expand all Collapse all HELSINGIN YLIOPISTO Finland EU contribution No data Address YLIOPISTONKATU 4 00014 HELSINGIN YLIOPISTO See on map Total cost No data KATHOLIEKE UNIVERSITEIT LEUVEN Belgium EU contribution No data Address OUDE MARKT 13 3000 LEUVEN See on map Total cost No data MAX-PLANCK GESELLSCHAFT ZUR FOERDERUNG DER WISSENSCHAFTEN E.V. Germany EU contribution No data Address HOFGARTENSTRASSE 8 80539 MUENCHEN See on map Total cost No data THE CHANCELLOR, MASTERS AND SCHOLARS OF THE UNIVERSITY OF OXFORD United Kingdom EU contribution No data Address UNIVERSITY OFFICES, WELLINGTON SQUARE OX1 2JD OXFORD See on map Total cost No data
