ISO/IEC JTC1/SC29/WG11N1350
MPEG96
July 1996 / Tampere
| Source: | Leonardo Chiariglione - Convenor | |
| Title: | Tampere Press Release | |
| Status: | Approved at 35th WG11 meeting |
Tampere Press Release
Highlights
At its 35th meeting in Tampere (Finland), from 8-12 July 1996, hosted by the Finnish Standards Association (SFS), and organized by Nokia Corporation, the MPEG working group made good progress with the MPEG-4 standard. New models of the Video, Audio and Systems parts were drafted. Considerable progress was made on the multiplex. Although MPEG meetings last a week, it was almost too short to process all the 150 contributions to the Video part of the standard.
Two parts of MPEG-2 (RTI and DSM-CC, see below) moved to International Standard at this meeting.
The first results of informal experiments on the comparison of the extension of MPEG-2 Audio to the new, 'Non Backward Compatible mode' (NBC) have been reported. These show a significant saving in bitrate for the same audio quality. Therefore, excellent performance of NBC is expected in the formal comparative assessment which is scheduled to be reported at the November 1996 meeting. NBC has successfully reached the stage of Committee Draft of the International Standard.
MPEG has created its own homepage on the Internet, providing information on the what, where, when and how of the MPEG standards. The homepage, which is updated regularly, holds information about the standards and sections on 'Frequently Asked Questions', about for instance MPEG Audio and Video. The address is http://www.cselt.it/mpeg
Background
The work in MPEG is carried out in several sub groups, that work on specific issues. Over 300 experts took part in the work in Tampere, working in several sub groups on specific issues. The next meeting will take place from 30 September through 2 October in Chicago, United States, hosted by Motorola. This will be an extra meeting, made necessary by the rapid progress in the 'Verification Models'. These VMs are the standards-to-be for MPEG-4 Audio, Video, and the MPEG-4 Systems and Description Language. MPEG-4 is currently the focus of MPEG's activity. MPEG-4 will be standard for the coded representation of audiovisual objects, that can be delivered through a variety of networks, and allow interactive use and re-use. A video object is for instance a dog, and audio object could be the sounds produced by that dog. Also computer-generated objects are important to MPEG-4.
A detailed list of the results of the 35th meeting follows below. The list is largely organized according to the several sub groups that MPEG knows.
Details
MSDL
The activities in developing the MSDL (MPEG-4 System and Description Languages, the 'system layer' of MPEG-4) were focused on defining how to multiplex the audio, visual, audiovisual and other objects into a single bitstream. This multiplex also carries information on the relationship between these objects, that is necessary to present them to a user, e.g. on a screen or through loudspeakers, in a synchronized way. Work was also done on defining how MPEG-4 decoding and composition tools can be used in a flexible environment, such as a PC. In this case the complete coding scheme is not fixed, but it can be configured according to specific user or application needs. To this end, API's (Application Program Interfaces) will be defined for these tools.
Next to listing information on the MPEG homepage, the MSDL has its own homepage:
http://www-elec.enst.fr/msdl/
Synthetic Audio and Video
An important property of MPEG-4 is the integration of natural and synthetic audio and video. The work in the synthetic area concentrated on the 'Call for Proposals', in which MPEG asks for methods to efficiently represent synthetic audio and video objects.
The first issue that this group will deal with is the coding of a talking head and the generation of artificial speech and audio. The next step is to add interactivity, and expanding the complexity of the virtual environment. The homepage of this group is http://www.es.com/mpeg4-snhc
Audio
Next to the work on the Non-Backward Compatible mode for
MPEG-2, the audio experts have defined the requirements for
MPEG-4. These requirements define the expectations of the experts
in MPEG in relation to the various applications to which MPEG-4
is targeted. The presence of clear requirements is essential in
the development of meaningful software and hardware embodiments.
This exercise was also carried out in the other groups developing
the MPEG-4 standard.
Video
The video group issued its third version of their Verification
Model. One of the most important capabilities supported by MPEG-4
Video is coding video objects with arbitrary shape. The
representation of these objects' shapes can now be done twice as
efficiently, and an even larger step is expected for next
meeting. It is possible to send or store these visual objects
with different qualities: using the same coded representation,
decoding to a basic quality can be used by simple systems or on
low capacity channels, but also high quality decoding is
possible. To draft the new VM, over 150 contributions on possible
improvements were evaluated.
MPEG-2
The Digital Storage Media - Command and Control (DSM-CC) part of the MPEG-2 standard progressed to International Standard, and the work on DSM-CC is now finished. DSM-CC provides the standard interface and API for video servers and set top boxes, that is being widely implemented, and will move the industry to a new level of inter-operability.
Also the Real-Time Interface (RTI) part of MPEG-2 progressed
to International Standard. The RTI facilitates inter-operability
by defining timing constraints on the real-time delivery of
MPEG-2 Tranport Streams.
Test
In July '97, the current status of the MPEG-4 standard will be
thoroughly tested against the requirements for the standard, and
other technology available in the market place, to ensure that
the developments are on the right track. The test group has
started to draft the outline for these tests. These tests will
pose a challenge in themselves, as requirements will be evaluated
that were never tested before on this scale, such as the
capability to cope with errors in mobile networks, or the quality
of individual objects in an audiovisual scene.
New subgroup organisation
At Tampere a new subgroup organisation was implemented, with the creation of a Requirements group, the transfer of MSDL activity to the Systems group and the creation of an SNHC group and a Liaison Group.
In a departure from its normal practice which does not mention
individual contribution in press releases, WG11 expresses its
gratitude and appreciation to Dr. Cliff Reader of Samsung for his
long standing leadership and contributions to WG11 specifically
and to worldwide standardization in general. Dr. Reader has
played a major role in the 3 generations of Audio/Video coding
standards developed by MPEG. The current efforts underway in MPEG
4 have felt his presence and inspiration since it was approved in
1993. We will miss him and it is our hope that some day he will
be able to find his way back to us. We wish Cliff the best of
success in all future endeavors.
For further information, please contact:
Dr. Leonardo Chiariglione, (Convenor of MPEG)
CSELT
Via G. Reiss Romoli, 274
10148 Torino, ITALY
Tel.: +39 11 228 6120;
Fax: +39 11 228 6299
Email:
leonardo.chiariglione@cselt.it