[M4IF Technotes] RE: Beware

[Stephan Wenger]

Dear Brian,
thanks for letting me know.
I'm no more active in VQEG, and certainly have not authorized the 
deployment of the few MPEG sequences VQEG uses.  But I can say that
a) Most of the VQEG sequences were donated by individual firms and are not 
the usual MPEG sequences.  In fact, VQEG uses many more sequences than the 
MPEG sequences, and the MPEG sequences were primarily used as anchor 
material, because their properties are very well known.
b) There are a number of sequences on the VQEG site that look like the MPEG 
sequences, but are not identical to the material on certain CDs with MPEG 
labels I got my hands on.  This is particularly true for Flower 
Garden.  Please check this before you jump to conclusions.
c) Many of the members of VQEG were (and some still are) active in MPEG as 
well, or work for companies that are active in MPEG.  I don't know whether 
the license the original contributors gave to MPEG was an exclusive 
license.  But if not, it would probably be ok to donate them to another 
standardization body as well, which has its own (less restrictive) rules 
regarding the public availability of material.
d) All MPEG sequences are widely (though probably illegally) used in 
academia and for research purposes.
e) The right people to talk to on this issue are the chairs of VQEG, Philip 
and Arthur.  Their coordinates can be found on the VQEG home page, located 
at www.vqeg.org.
As a personal comment, and off the record I would like to add that trying 
to take these sequences out of the public domain (where they are de-facto, 
though apparently not de-jure) would not only potentially harm the VQEG 
project, but also the development of video technology itself.  And it would 
be practically unfeasible.  Every university department working on video 
coding I'm aware of has those sequences.
Best regards,
Stephan
At 06:30 PM 7/23/2002 -0700, Brian Dipert wrote:
>Followup; after visting Stephan's site at
>http://kbs.cs.tu-berlin.de/~stewe/vceg/, I may be confused, but I still
>think not. If you visit www.vqeg.org (mirrored at http://media.xiph.org)
>you'll find that the sequences you compressed AREN'T there. The VQEG
>sequences are D1 (both PAL and NTSC) resolution and have file names of style
>srcXX_ref__X25.yuv....I also used these in my project, and they're public.
>Based on the names and resolutions of your files, I think they're actually
>MPEG-4 standards committee material, which as I mention below is NOT public.
>I've copied Stephan on this email and perhaps he can clarify
>==============================
>Brian Dipert
>Technical Editor: Mass Storage, Memory, Multimedia, PC Core Logic and
>Peripherals, and Programmable Logic
>EDN Magazine: http://www.edn.com
>Contributing Editor, CommVerge Magazine: http://www.commvergemag.com
>1864 52nd Street
>Sacramento, CA   95819
>(916) 454-5242 (voice), (916) 454-5101 (fax)
>mailto:bdipert   edn.com
>Visit me at http://www.bdipert.com
>
> > -----Original Message-----
> > From: Brian Dipert [mailto:bdipert   pacbell.net]
> > Sent: Tuesday, July 23, 2023 6:24 PM
> > To: singer   apple.com
> > Cc: Rob Koenen; technotes   lists.m4if.org
> > Subject: Beware
> >
> >
> > Umm...David. I also used the University of Hannover/MPEG-4 test
> > sequences in my video compression hands-on project, due to go
> > live Thursday morning at www.edn.com (any and all feedback
> > welcomed). Leonardo Chiariglione (yes, THAT Leonardo) from the
> > MPEG standards group was VERY emphatic (Rob'll back me up on
> > this) that I NOT post either the original YUVs or any lossless-
> > or even lossy-compressed variants of them to a public website,
> > since (as you point out on your
> > http://index.apple.com/~singer/sequences/testseq.html website)
> > they're copyright-protected. Note that the area of the Hannover
> > FTP site where the originals reside is password-protected; only
> > for use by MPEG members. Just thought I'd let you know that an
> > angry Italian may soon be headed your way. Conversely, if you
> > secure permission rights......please let me know, as I'd like to
> > post my stuff too!
> > ==============================
> > Brian Dipert
> > Technical Editor: Mass Storage, Memory, Multimedia, PC Core Logic
> > and Peripherals, and Programmable Logic
> > EDN Magazine: http://www.edn.com
> > Contributing Editor, CommVerge Magazine: http://www.commvergemag.com
> > 1864 52nd Street
> > Sacramento, CA   95819
> > (916) 454-5242 (voice), (916) 454-5101 (fax)
> > mailto:bdipert   edn.com
> > Visit me at http://www.bdipert.com