Inter-media
Semantic
Extraction &
REasoning

 

Presentation
Objectives
Partners
Activities
  + Meetings
  + Exchanges
  + Materials

 

 

 

The first year will be mostly dedicated to the definition of this common modelling based on each team past experiences. We will also take into account the basic extractions tools developed in each teams. We will then propose a framework  for the common representation.

  • 4 meetings in Asia, one in Japan, one in Singapore one in Taiwan and one in Vietnam
  • preparation and collection of data
  • description of existing tools
  • proposition of a common model
  • first analysis tools for data
  • beginning (1st year) of join PhD

Among the interesting media analyser, we propose:

  • Modelling human body for tracking in video signal

  • spoken language transcription and keywords extraction

  • noise and sound analyzer

  • video analysis for extracting useful context information

The second year will be the first definition of the common multimedia reasoning toolkit, and some domain specific experiences. At the end of this year, one has to be ale to defined what knowledge remain domain specific, and what can be move to the common modelling.

The third year will be devoted to the finalisation of the common model and by the set up of a common experimentation on a sub part of the data sources to be defined.

Page updated February 9th, 2005