Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!emory!ogicse!milton!woobin
From: woobin@milton.u.washington.edu (Woobin Lee)
Newsgroups: comp.ivideodisc
Subject: Re: JPEG/MPEG standard development
Message-ID: <12986@milton.u.washington.edu>
Date: 14 Dec 90 05:38:13 GMT
References: <9312@pasteur.Berkeley.EDU>
Organization: University of Washington, Seattle
Lines: 49

In article <9312@pasteur.Berkeley.EDU> larry@postgres.UUCP (Larry Rowe) writes:
>
>Second, MPEG -- the motion video standard.  This committee is much further
>behind the JPEG committee.  They are currently working on a draft document,
>but nobody outside the committee has seen the current drafts.  

I'm not in the committee, but I have a copy of MPEG draft.
I don't think it's confidential, as I didn't sign anything to get it.

>Supposedly,
>the specification will have three parts: video, audio, and system.  The
>system part specifies how video and audio packets are combined.  At present,
>a draft exists for the video part, an outline exists for the audio part, and
>nothing has been written for the system part.  So, it will be quite a while
>before we see a complete specification.  

I don't know much about audio, but from the contents of the draft material,
I think the audio part is closest to the finalization.

>Touchtone said that the video part
>was based on a CCITT standard (ptime 64?) rather than the JPEG standard.
>This is interesting because the video compression hardware that I'm familiar
>with is based on JPEG (e.g., Next video board).

MPEG standard is very similar to CCITT, such as in motion estimation, 
intra/non-intra decision, DCT/IDCT, and quantization schemes. One major
difference is the addition of Bidirectionally Predicted frames in MPEG
to Intra and Predicted frames of CCITT.
The MPEG standard is intented for full-motion video with storage/transfer media 
capable of supporting 1.5 Mbps (CD ROM, DAT, HD etc).
The NeXT board incorporating C-Cube chip is intended for still image 
compression. The JPEG standard is for still images, and therefore does not
exploit temporal redundancy as in MPEG or CCITT.

>I'm going to try to follow what's happening on these committees, so I'll
>keep you informed.
>	Larry Rowe

I think the committe meets once every two months, and the last meeting was
(or will be?) held in Berlin. I haven't got any report from this meeting yet.

By the way, does anybody know where I can get a copy of CCITT H.261 standard
and their reference model (RM8)?

Woobin Lee -----------------------------------------------------------------
				||
Image Computing System Lab	||	woobin@u.washington.edu
University of Washington	||
Seattle, WA 98195		||	(206) 543 - 1017