4M, tools, Algorithm Ontology UC

Dear all

in the following our comments to the emails received (Raphael, Zeljko
and Jeff).


- 4M Infrastructure (23rd march Raphael's email)
4M is an infrastrucure developed by ISTI-CNR in the frame of the EU
MUSCLE Network of Excellence.
Some information about 4M can be found at
http://image.ntua.gr/swamm2006/resources/paper01.pdf
At present, 4M extracts mpeg-7 features from audio, images and videos.
The software has been implemented in Java (we did not use XM (C
language) because it has limits on media management). Multimedia
extractor is based on Vizir (that we have extended) and mpeg7audioencoder.
We also extended eXist (XML native database) in order to better handle
users' groups (our improvements have been sent to the developers: the next
version of eXist will contain our modifications).
4M also performs some multimedia similarity based on the extracted features.
A prototype of an image annotation module is now available.
Work to improve 4M is still in progress.
At present, only MUSCLE-NoE partners are allowed to access and use 4M.


- Links between tools and usecases (8th april Zeljko's email)
Mpeg7Audio Encoder:
http://www.ient.rwth-aachen.de/team/crysandt/software/mpeg7audioenc/
Vizir: http://vizir.ims.tuwien.ac.at
eXist: http://exist.sourceforge.net/
In attachment the DOAP file of 4M.

- Algorithm Usecase (9th april Jeff's review email )
we have added an "interoperability aspects" separate section on the wiki.


Again, thank you all !

Massimo & C. ;-)

Received on Thursday, 26 April 2007 14:05:03 UTC