Project details


Centrum pro multi-modální interpretaci dat velkého rozsahu

Project Identification:GBP103/12/G084 (CEP Id: GBP103/12/G084)
Project Period:1/2012 - 12/2018
Investor:link to a new windowCzech Science Foundation
Programme / Project Type:Projects to promote excellence in basic research -
MU Faculty/Unit:
Faculty of Informatics
MU Investigator:Prof. Ing. Pavel Zezula, CSc.
Project Team Member:RNDr. Michal Batko, Ph.D.
Assoc. Prof. RNDr. Vlastislav Dohnal, Ph.D.
RNDr. David Novák, Ph.D.
Cooperating Organization:
link to a new windowCzech Technical University Prague
Responsible Person:Prof. Ing. Jiří Matas, Ph.D.
link to a new windowCharles University Prague
Responsible Person:RNDr. Pavel Pecina, Ph.D.
link to a new windowUniversity of West Bohemia in Pilsen
Responsible Person:Prof. Ing. Josef Psutka, CSc.

The project aims at exploiting large collections of unlabeled multi-modal data, mainly video footage, to further state-of-the-art in video, audio and natural language understanding, interpretation, annotation and retrieval by combining unsupervised and semi-supervised learning. It will address problems that are very difficult (some probably impossible) to solve in a single modality by adopting an interdisciplinary approach. Progress in individual research areas - vision, natural language processing and speech will be achieved by co-training and by exploiting results of other modalities as cross-training data. For efficient processing of large data collections appearing in all modalities, the project will also concern generic problems of organization, indexing, and searching based on similarity that is fundamental for building real-life applications.