Skip to content

Sections
Personal tools
You are here: Home » Research » Projects » Phase II » IM2.MCA ( Multimodal content abstraction )
Address
NCCR IM2
c/o Idiap
Centre du Parc
Rue Marconi 19
Case Postale 592
CH-1920 Martigny
Switzerland

tel. +41 27 721 77 11
fax +41 27 721 77 12


 

IM2.MCA ( Multimodal content abstraction )

Document Actions

Multimodal content abstraction

 

IM2.MCA
IP Head: Stephane Marchand-Maillet (CVML/UniGE)
Partners: CVML/UniGE, ISSCO/UniGE, IDIAP.
IM2.MCA addresses the issues of content-based indexing, retrieval, semantic analysis and mining of multimedia data based on multiple modalities. As such, it focuses on all activities related to content abstraction of mono- and multi-modal information, based on the annotation and features of all signal-based documents (i.e. documents representing raw signals.
Such multimedia documents include video, audio (inc. speech), images, slides, complex printed documents, handwritten notes as instances of containers of spatial, temporal and multimodal information.

The main objectives of the IP are:
  • to enrich multimedia documents, such as dialogue recordings and associated static documents, with cross-modal metadata, either via fully automatic recognizers or via computer-aided manual methods;
  • to extract and mine information from the above documents and associated metadata;
  • to define and implement multimodal indexing and retrieval approaches for the above information content;
  • to define and implement strategies for managing the above information content at the collection level.

Last modified 2006-02-03 15:37
 
 
Powered by Plone