Re: [MMSEM-UC] Low-level feature extraction use case

Dear Raphael,

My aim is to develop upon a UC that will identify the paths
towards automating semantics extraction from multimedia. 
In particular, to report on corresponding interoperability problems (both syntactic and semantic)
and ideally suggest possible solutions using SW technology.
My development will be mainly viewed from the multimedia analysis perspective
and it is certainly needed support from other people not only from this perspective
but also from the pure SW technology point of view.
I believe that the suggestion of Susanne can be very well situated to the above view.
A collaboration with her would be more than welcome.

I'll try to give a concrete description by tomorrow (before telecon)
as to enable all interested participants to express their corresponding opinion. 

Till tomorrow,
best regards,


Ioannis PRATIKAKIS, Dipl. Eng., Ph.D
Research Scientist
Computational Intelligence Laboratory (
Institute of Informatics and Telecommunications
National Center for Scientific Research "Demokritos"
P.O. BOX   60228
GR-153 10 Agia Paraskevi, Athens, Greece.
Tel:     +30-210-650 3183
Fax:    +30-210-653 2175

----- Original Message ----- 
From: "RaphaŽl Troncy" <>
To: "Susanne Boll" <>; <>
Cc: "MMSem public" <>
Sent: Wednesday, October 11, 2006 10:51 AM
Subject: [MMSEM-UC] Low-level feature extraction use case

Dear Susanne and Ioannis,

> I could not attend the last Telco but I could rework the
> low level feature extraction use case to make it suitable
> to the desired structure and prepare it for a later review.

Great, excellent, thanks Susanne !
Ioannis, you were the one who originally proposed this use case, during the
Amsterdam F2F meeting. I have put on the wikis your original thoughts ...
The use case is now different with the view of Susanne. Do you share this
view ? Would you like to work together with Susanne ? Please speak up !

> However, my interest would be a combination of tagging
> and low level features (mainly) for images. I can make it a
> photo low level feature use case and we can see what the
> other think. So my use case would be targeted in the field
> of

I think that could make an ideal use case and I strongly encourage you !
Actually, I was looking for a use case that could take back the work done
around rdfpic and this note in particular :-)

Best regards.


RaphaŽl Troncy
CWI (Centre for Mathematics and Computer Science),
Kruislaan 413, 1098 SJ Amsterdam, The Netherlands
e-mail: &
Tel: +31 (0)20 - 592 4093
Fax: +31 (0)20 - 592 4312

Received on Wednesday, 11 October 2006 09:08:09 UTC