W3C home > Mailing lists > Public > public-xg-mami@w3.org > February 2011

Minutes of 28-Jan-2011 F2F meeting

From: Nobuhisa Shiraishi <n-shiraishi@bq.jp.nec.com>
Date: Sat, 19 Feb 2011 18:32:18 +0900
To: public-xg-mami@w3.org
Message-Id: <20110219183143.61AF.C4E44556@bq.jp.nec.com>
Dear MAMI-XG members,

Attached please find the minutes of our 1st F2F meeting on 28 Jan.

With my best regards,
Nobu.

------------------------------------------------------------
= MAMI-XG F2F Meeting Minutes =

* Date: 28 January 2011
* Time: 15:00-16:00 JST (GMT+9)
* Place: NEC Tamagawa Plant(Japan) N302 meeting room

== Attendees ==
* Present
Masuo(NTT), Tajiri(Fujitsu - substitute), Yamada(NEC), Shiraishi(NEC)
* Observers
Inagaki(NTT), Hashimoto(NEC)
* Regrets
Matsukura(Fujitsu), Kamei(NEC)
* Chair
Shiraishi(NEC)

== Summary ==
* Clarify the strong points of MAMI
- Survey existing specifications of service IFs and analyzer IFs.
* List up the use cases of MAMI
- Organize the use cases of MAMI by the types of retrieved data 
and by the types of operations to the retrieved data.

== Action Items ==
(1) Survey the existing specifications of service IFs.(NEC)
(2) Survey the existing specifications of analyzer IFs.(NEC)
(3) List up and organize the use cases of MAMI by the type of objects 
retrieved from video images and by the types of operations to the retrieved
objects(NEC)

The works for above action items will be shared and discussed on the
mailing list.

== Next meeting ==
- Next F2F meeting is Scheduled as following:
Date: 10 Mar 2011
Time: 10:00-12:00 JST (GMT+9)
Place: NEC Tamagawa Plant(Japan) N311 meeting room

== Discussions ==
(TM: Masuo, HI: Inagaki, OH: Hashimoto, NS: Shiraishi)

=== 1. Goals and outputs of our XG ===
* The user analysis, use case analysis, and organization of them will be
 outputs of MAMI?(TM)
-> Yes. These discussions are very important and essential.
We should publish and record the result of these discussions.(NS)

=== 2. Analysis of MAMI users ===
* We should clarify the merit of switching to MAMI.(HI)
-> I think MAMI will not the replacement of existing analyzer IFs.
Analyzers may adapt to MAMI supporting the existing legacy IFs. (NS) 
-> We should clarify the merit of supporting both MAMI and legacy IFs.
We should also consider the business aspects. (HI)
-> The standardization of analyzer IFs have not discussed enough
compared to the standardization of service IFs. Analyzers have provided 
their proprietary IFs and have required to introduce their proprietary
drivers. (OH)
-> Which will be our main target, existing analyzers or new analyzers?
Our analysis and strategy depend on our target. (HI)

* Avoid exhausting our discussion, we had better to limit the scope our
target to the direct users of MAMI. For example, sensor vendors will be
out of scope of our discussion. (HI)
-> I agree with you. We had better to take Web browser vendors into our
scope, because the some services will be provided on Web browser. (NS)
-> Clarification of our target is very important. We should clarify our
target before starting our discussion. (HI)
-> We conclude that sensor vendors are out of our scope and web browser
vendors are inside of our scope. (OH)
-> NEC will survey the existing analyser IFs. (OH)

* Is the privacy of analyzed data discussed well in other working groups?
For example, in MFWG, who is the owner of the fragment data of video
images, analyzer or the owner of the original video? (TM)
-> I think it is not discussed in MFWG in these months. I have not checked
previous discussions. (NS)
-> The business of metadata is difficult, particulary metadata which
includes privacy data. General consumer is much severer than companies. (OH)
-> The owner of analyzed data is unclear. There still remain various
problems in the businesses which use analyzed data for various services. (OH)

=== 3. Organizing and rating the use cases of MAMI ===
* Is there any good examples of organizations of use cases in other
working groups? (OH)
-> In other working group, we listed up the functions we provide at
first. Next, we discussed various services which will be enabled by the
functions. (HI)
-> We can take various approaches for studying the use cases of MAMI.
I am not sure which will be the best way. (HI)

* How can we rate the use cases of MAMI?(HI)
-> We might be able to rate the use cases from the viewpoint of businesses. (OH)

* How about starting with listing up the use cases of video analysis? (OH)
-> We will be able to organize them by the types of objects retrieved
from video and by the types of operations to the retrieved objects. (OH)
-> After that, we might be able to rate them from the viewpoint of
businesses. (OH)

* How was the approach of use case analyses in FG-Smart in ITU-T? (HI)
-> In FG-Smart, there were several categories of use cases already defined.
The discussion in FG-Smart was requirement analysis rather than use case
analysis. (TM)

* For the use case analysis, which approach will be better, bottom-up
(from functions analysis) or top-down (from user needs)? (OH)
-> If we can clarify the user needs, the top-down approach will be better. (OH)

* What is the purpose of our use cases analysis? (OH)
-> List up the requirements of MAMI based on the use case analysis, and
refine the technical specification of MAMI based on the requirements
analysis. (NS)

* How will we collaborate with MFWG on the use case discussion?(OH)
-> I have already input 3 use cases to MFWG. If there are other use
cases which are relevant to MFWG, I will input them to MFWG. (NS)

* Let's list up the use cases of MAMI organizing by the types of
retrieved data and by the types of operations to the retrieved data.
As a starting point, let's focus on video analysis and think about their
general use cases, not limited to business companies but including
general consumers. (OH)
------------------------------------------------------------

-----
Nobuhisa Shiraishi <7M3RLA/1>
Information and Media Processing Laboratories, NEC Corporation.
E-mail:n-shiraishi@bq.jp.nec.com Tel:+81-44-431-7672 Fax:+81-44-431-7588
Received on Saturday, 19 February 2011 09:33:08 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 19 February 2011 09:33:08 GMT