|
Project Name: Multimedia Indexing and Searching Environment
Abbreviation: MUMIS Start date:
July 1, 2000 End date:
May 1, 2003 Project Description:
MUMIS will develop basic technology for the automatic indexing of
multimedia broadcast content, in the following stages:
- Use data from different media sources (documents, radio and
television programmes) to build a specialised set of lexica and an
ontology for the selected domain (soccer).
- Digitise non-text data and apply speech recognition techniques
to
extract text.
- Develop information extraction techniques and apply them to the
texts to extract significant text elements (such as the names of
players in a team, goals scored...) and use these to build
annotations
- Build a novel type of merging tool, to combine the event
descriptions generated from different data sources, and from sources
in different languages (Dutch, English, German).
- Store the merged event descriptions (the domain ontology) in a
central database, with relevant metadata.
- Develop a user interface (in Dutch, English, German and Swedish)
to enable professional users to query the database, by selecting from
menus of event descriptions, metadata, and so on, and to view video
fragments retrieved to satisfy the query.
Official Website
http://hmi.ewi.utwente.nl/Projects/mumis/
Project Funding
MUMIS is a project within the 5th Framework Programme IST
(Information Society Technologies) of the European Union.
Related Projects
- Twenty-One
- Pop-Eye
- Olive
- Druid
|
The following HMI-member(s) is/are coordinator of this Project
Franciska de Jong
Here you can find the publications
|
