ISO/IEC JTC 1/SC 29 N 1944
DATE: 1997-03-06
OUR REF.: 29CL373/29F046/29D126/
29N1944c.htm
29N1944x.gif (x:1-2)


ISO/IEC JTC 1/SC 29

Coding of Audio, Picture, Multimedia and Hypermedia Information

Secretariat: Japan (JISC)

DOC. TYPE Meeting Report
TITLE Meeting Report, the Thirty-seventh ISO/IEC JTC 1/SC 29/WG 11 Meeting, 1996-11-18/22, Maceio, Brazil [JTC 1/SC 29/WG 11 N 1415]
SOURCE Convener, ISO/IEC JTC 1/SC 29/WG 11
PROJECT NO. --
STATUS Draft meeting report
REFERENCES  SC 29 N 1851 (SC 29/WG 11 N 1413): Resolutions
ACTION ID. FYI
REQUESTED ACT. For SC 29's information
DUE DATE --
DISTRIBUTION P-, O- and L-members, ISO/IEC JTC 1/SC 29
Officers, ISO/IEC JTC 1/SC 29
Secretariat, ISO/IEC JTC 1
ISO/IEC ITTF
MEDIUM Def/D
NO. OF PAGES N/A



Narumi Hirose, Secretariat, ISO/IEC JTC 1/SC 29
IPSJ/ITSCJ*, Room 308-3, Kikai-Shinko-Kaikan Bldg., 3-5-8, Shiba-Koen, Minato-Ku Tokyo 105 Japan
Telephone: +81-3-3431-2808; Facsimile: +81-3-3431-6493;
Telex: 2425340 IPSJ J; E-mail: hirose@itscj.ipsj.or.jp
* Information Processing Society of Japan/Information Technology Standards Commission of Japan (A standards organization accredited by JISC)



INTERNATIONAL ORGANISATION FOR STANDARDISATION

ORGANISATION INTERNATIONALE DE NORMALISATION

ISO/IEC JTC1/SC29/WG11

CODING OF MOVING PICTURES AND AUDIO

ISO/IEC JTC1/SC29/WG11 N1415

November 1996

Source:Leonardo Chiariglione - Convenor
Title:Report of 37th WG11 meeting
Status:

1. Opening

The 37th WG11 meeting was held in Maceio;, AL, BR on 17-21 November at the kind invitation of Associacao Brasileira de Normas Tecnicas (ABNT), the Brazilian National Body. Mr. Raul Colcher, the Chairman of the Brazilian JTC1 committee welcomed the delegates.

2. Roll call of participants

The list of participants is given in Annex 1.

3. Approval of agenda

The agenda given in Annex 2 was approved.

4. Allocation of contributions

Annex 3 gives the list of input contributions to the meeting.

5. Communications from Convenor

There was no specific communication made.

6. Report of previous meeting

This was approved.

7. Processing of NB Position Papers

These were considered and a reply produced and agreed.

8. MPEG Phase 2

8.1 Audio

No activity related to 13818-3 took place.

8.2 Verification of MPEG-2

The meeting learnt with great enthusiasm about the very high performance demonstrated by the NBC standard, that nearly halved the bitrate required, for a comparable quality, compared to MPEG-1 Audio.

8.3 Amendments

MPEG-2 Video amendment 4 was progressed to DAM status.

8.4 Part 6 (DSM-CC)

No activity took place on this topic.

8.5 Part 7 (NBC Audio)

NBC Audio was promoted to DIS. The acronym was changed from NBC to AAC (Advanced Audio Coding).

8.6 Part 10 (DSM-CC Conformance)

This being the first meeting of the DSM group after completion of part 6, it was possible to plan for work related to part 10. Chris Adams, who had been served for a long time as Chairman of the DSM-CC group resigned. Vahe Balabanian was appointed in his stead.

8.7 Workplan

This was approved.

9. MPEG Phase 4

9.1 Requirements

A new version of the requirements document was approved.

9.2 Tools

Tools was systems and synthetic and natural audio and video coding were discussed and results of core experiments analysed.

9.3 Verification Models

Systems, Audio, Video and SNHC were approved.

9.4 Tests

A test procedure document for the July '97 tests was approved.

9.5 Call for proposals

A call for proposals for competitive tests in July 1997 was approved.

9.6 Simulation software

The Systems and SNHC groups agreed to make available encoder and decoder software. The Audio and Video groups agreed to make decoder software available.

9.7 Working Draft

The Systems group produced a new version of the WD. The Video and Audio groups produced a first version of their WDs. The SNHC group will produce a WD at the April meeting, but this will be folded into the Systems, Video and Audio WDs, respectively.

9.8 Workplan

This was approved.

10. MPEG Phase 7

A special session, organised by the Requirements group, produced a "Context and Objectives" document.

11. Overall WG11 workplan

This was approved.

12. Liaison matters

Input documents were analysed and liaison letters, in some cases with attached WG11 documents, were produced.

13. Administrative matters

13.1 Schedule of future MPEG meetings

The 5-year meeting schedule was revised extending supported meeting until November 1998.

13.2 Promotion of MPEG

The Audio groups produced an "Audio FAQs" document for the MPEG home page. The other groups were requested to produce similar documents.

14. Organisation of this meeting

14.1 Tasks for subgroups

The following tasks were assigned to the subgroups:

Requirements

Systems

Video

Audio

SNHC

Test

DSM

Subgroup reports are given in the following annexes:
RequirementsAnnex 4
SystemsAnnex 5
VideoAnnex 6
AudioAnnex 7
SNHCAnnex 8
TestAnnex 9
DSMAnnex 10
LiaisonAnnex 11

14.2 Finalisation of meeting allocation

In addition to regular subgroup sessions, the following joint meetings were held.
Joint meeting betweenDay Time
SNHC-Systems Mon 15:00-16:00
Audio-Test (NBC)Mon 15:30-16:30
DSM-Systems-RequirementsMon 16:00-17:00
Video-Systems Mon 17:30-19:00
Video-Test Tue09:00-10:00
Audio-Implementation Tue 11:00-12:00
SNHC-Requirements Tue 17:00-18:00
Test-RequirementsWed 14:30-15:30
Test-AudioWed15:30-16:30
SNHC-Systems Wed15:00-16:00
Requirements-AudioThu 09:00-10:00
Test-Implementation-Video Thu 10:00-10:30
Video-Implementation Thu 14:00-14:30
Video-Requirements Thu 14:30-15:00
Systems-Requirements Thu 17:00-18:00
Requirements-SystemsFri 09:00-10:00

15. Planning of future activities

The following ad-hoc groups were established.
1498Adhoc group on Computational graceful degradation
1479Ad-hoc group on Core experiments on coding efficiency in MPEG-4 video
1476Ad-hoc group on Core experiments on coding of arbitrarily shaped objects in MPEG-4 video
1477Ad-hoc group on Core experiments on error resilience in MPEG-4 video
1478Ad-hoc group on Core experiments on multifunctional coding in MPEG-4 video
1435Adhoc group on editing of DIS 13818-7 (NBC Audio)
1501Adhoc group on editing of SNHC VM version 2.0
1480Ad-hoc group on Editing the Documents of the MPEG-4 Video Verification Model and the MPEG-4 Video Working Draft
1448Adhoc group on face and body animation
1491Adhoc group on IPR issues in MPEG-4
1445Adhoc group on July 1997 tests
1497Adhoc group on low delay issues
1446Adhoc group on Media integration of text and graphics
1440Adhoc group on MPEG audio FAQ
1437Adhoc group on MPEG-2 NBC conformance and technical report preparation
1439Adhoc group on MPEG-4 core experiments
1496Adhoc group on MPEG-4 requirements and profiles
1490Adhoc group on MPEG-4 signaling
1438Adhoc group on MPEG-4 WD and VMs
1436Adhoc group on NBC complexity reduction
1447Adhoc group on SNHC Audio
1449Adhoc group on SNHC texture coding
1489Adhoc group on Systems architecture evolution
1488Adhoc group on Systems multiplex specification and signaling
1487Adhoc group on Systems VM
1486Adhoc group on Systems WD editing
1506Adhoc group on Video VM padding complexity

16. Resolutions of this meeting

These were approved.

17. A.O.B

There were no other businesses.

18. Closing

The meeting closed at 97/11/22 21:45.


Annex 1

List of participants
JOHN ARNOLDUNIVERSITY OF NEW SOUTH WALES AU
MICHAEL FRATERUNIVERSITY OF NEW SOUTH WALES AU
BERT DEKNUYDTESAT - KULEUVEN BE
FILIP DEFOORTESAT - KULEUVEN BE
GAUTHIER LAFRUITIMEC BE
THIERRY DELMOTU.C.L.TELECOMMUNICATION LAB. BE
YVES CHRISTOPHEVRIJE UNIVERSITEIT BRUSSEL BE
RAUL COLCHERABNT/CB-21 BR
DEISE FREITASCLIP EVENTOS BR
RAFAEL LONTRACLIP EVENTOS BR
R A. B. GONÇALVESCLIP EVENTOS BR
A. C. FRANÇA PESSOACPqD - TELEBRÁS BR
CLÁUDIA COSTASTAFF BR
IVANA COÊLHOSTAFF BR
NEILA TOMÉSTAFF BR
CARLOS DE BRITO NOGUEIRATV GLOBO BR
EDUARDO ANTONIO B. SILVAUNIVERSIDADE FEDERAL DO RIO DE JANEIRO BR
GELSON VIEIRA MENDONÇAUNIVERSIDADE FEDERAL DO RIO DE JANEIRO BR
VAHE BALABANIANNORTHERN TELECOM CA
MARCO MATTAVELLIEPFL CH
FRED JORDANEPFLCH
EBRAHIMI TOURADJEPFL CH
WEN-YUAN CHENCOMPUTER & COMMUNICATION RESEARCH LABS. CN
YUNG-HUA HUNGCOMPUTER & COMMUNICATION RESEARCH LABS. CN
THOMAS STINGLFZI KARLSTUHE DE
KLAUS DIEPOLDBETA TECHNIK GMBH DE
GUNNAR NITSCHEBOSHDE
SVEN BAUERBOSHDE
ANDREAS GRAFFUNDERDEUTSCHE TELECOM AG DE
ANGELIKA KNOLLDEUTSCHE TELECOM AG DE
PETER LISTDEUTSCHE TELECOM AG DE
FRANK FEIGEDEUTSCHE TELEKOM AG DE
KARL - HEINZ BRANDENBURGFRAUNHOFER INSTITUT IIS-A DE
GERHARD ZIMMERFRAUNHOFER INSTITUTE DE
HERRE JURGUENFRAUNHOFER INSTITUTE DE
MARTIN DIETZFRAUNHOFER INSTITUTE DE
JAN DE LAMEILLIEUREHEINRICH HERTZ INSTITUT - HHI DE
THOMAS SIKORAHEINRICH-HERTZ INSTITUTE - HHI DE
GERARD STOLLINSTITUT FAR RURDFUNLETECHNICK DE
RAINER PÜLLENPULTITUDE COMPUTER ELECTRONICS DE
GERALD KNABEQ-TEAM DR. KNABE GEUBTT DE
FRITZ SEYTTERSIEMENS AG DE
CARSTEN HERPELTHOMSON MULTIMEDIA DE
JENS SPILLETHOMSON MULTIMEDIA DE
THOMAS STOCKHAMMERUNIVERSITY MUNICH DE
ANDREAS HUTTERUNIVERSITY MUNICH DE
PETER KUHNUNIVERSITY MUNICH DE
BERNHARD GRILLUNIVERSITY OF ERLANGEN DE
BERND EDLERUNIVERSITY OF HANNOVER DE
PETER GERKENUNIVERSITY OF HANNOVER DE
NOEL BRADYTELTECEI
LUIS TORRES URGELLPOLYTECHNIC UNIVERSITY OF CATALONIA UPC ES
MIGUEL ROSERTELEFÔNICA INVESTIGACION Y DESARROLLO ES
MARTA KARCZEWICZNOKIA FI
PASI OJALANOKIAFI
MAURI VAANANENNOKIA FI
MARCIN RZESZUIKONOKIA RESEARCH CENTER FI
ISABELLE AMONCUCANON FR
JEAN BARDAAFNORFR
JEAN-MICHEL BORDEAFNOR - FRENCH NATIONAL BODY FR
DOMINIQUE CURETCCETT FR
DOMINIQUE NASSECCETT FR
JULIEN SIGNESCCETTFR
LAURENT MAINARDCCETT FR
DOMINIQUE YONCISACFR
BRUNO LORETFRANCE TELECOM FR
GERARD EUDEFRANCE TELECOM FR
OLIVIER AVAROFRANCE TELECOM - CNET FR
JEAN-CLAUDE DUFOURDFRANCE TELECOM - ENST FR
GILLES PRIVATFRANCE TELECOM - CNET FR
DAVID MOLTERPHILIPS FR
ISABELLE CORSETPHILIPS FR
LIONEL BOUCHARDPHILIPS FR
CLAUDE ROLLINSGDL - SCAM FR
JEAN-FRANÇOIS VIALTHOMSON FR
AHARON GILLZORAN MICROELECTRONICS LTD. IL
VITTORIO BARONCINIF.U.B. IT
ROBERTO FLAIANIAETHRA IT
GIORGIO PARLADORIALCATEL IT
GUIDO FRANCESCHINICSELT IT
LAURA CONTINCSELTIT
LEONARDO CHIARIGLIONECSELT IT
MARCO GANDINICSELTIT
STEFANO BATTISTACSELT IT
ROBERTO POCKAJDIST - UNIVERSITY OF GENOVA IT
GIUSEPPE RUSSOFONDAZIONE UGO BORDONI IT
ITARU KANEKOASCIIJP
EISHI MORIMATSUFUJITSU LABORATORIES LTD. JP
YASUFUMI NAKAMURAFUJITSU LTD. JP
YASUO KATAYAMAGCLJP
YUICHIRO NAKAYAHITACHI LTD. JP
AKIRA DATEHITACHI, LTD. JP
KIKUCHI YOSHIHIROJAPAN JP
HIROAKI SUZUKIJVCJP
ICHIRO ANDOJVCJP
KOJI IMURAMATSUSHITA JP
KOJI YOSHIDAMATSUSHITA JP
NAOYA TANAKAMATSUSHITA JP
YUTAKA MACHIDAMATSUSHITA JP
TAKESHI NORIMATSUMATSUSHITA ELECTRIC INDUSTRIAL CO. JP
TAKUYO KOGUREMATSUSHITA ELECTRIC INDUSTRIAL CO. JP
CHOONG SENG BOONMATSUSHITA ELECTRIC INDUSTRIAL CO. JP
MINORU ETOHMATSUSHITA ELECTRIC INDUSTRIAL CO. JP
TAKANORI SENOHMATSUSHITA ELECTRIC INDUSTRIAL CO. JP
KOHTARO ASAIMITSUBISHI ELECTRIC CORPORATION JP
SHUN-ICHI SEKIGUCHIMITSUBISHI ELETRIC CORPORATION JP
MASAHIRO IWADARENEC CORP. JP
YOSHIRO MIYAMOTONEC CORP. JP
HITOSHI KOYAMANEC CORP. JP
TOSHIYUKI NOMURANEC CORP. JP
KAORU WATANABENHKJP
SHINICHI SAKAIDANHK JP
HIROYUKI FUKUCHINIPPON STEEL JP
KINYA OSANIPPON STEEL JP
HIROHISA JOZAWANTTJP
SATOSHI MIKINTTJP
TAKEHIRO MORIYANTTJP
NOBUHIKO NAKANTT DOCOMO JP
TOSHIO MIKINTT DOCOMO JP
SHINYA NAKAJIMANTTJP
NAKAI TOSHIHISAOKI ELECTRIC IND. CO. LTD JP
WU ZHIXIONGOKI ELECTRIC INDUSTRY CO., LTD JP
NEIL DAYRICOH CO. LTD. JP
HIROYUKI KATATASHARP CORPORATION JP
KEIICHI HIBISHARP CORPORATION JP
MASAMI OGATASONYJP
MASAYAKI NISHIGUCHISONY JP
TERUHIKO SUZUKISONY JP
YOICHI YAGASAKISONY JP
KAZUHISA HOSAKASONY JP
JUN MATSUMOTOSONYJP
KENZO AKAGIRISONYJP
TAKASHI KOIKESONYJP
YOSHIAKI OIKAWASONY JP
YUJI ITOHTEXAS INSTRUMENTS JP
TOSHIAKI WATANABETOSHIBA JP
JONGIL KIMDAEWOOKR
CHIETEUK ANNETRIKR
H. TEAM MAKEETRIKR
JANG DAE YOUNGETRIKR
MINSOO HAHNETRIKR
YOUNG-KWON LIMETRIKR
GWANG-HOON PARKHYUNDAI ELECTRONICS INDUSTRIES CO. KR
JOO-HEE MOONHYUNDAI ELECTRONICS INDUSTRIES CO. KR
HYUNWOOK PARKKAISTKR
JAE-WON CHUNGKAISTKR
SANG WOON SUHKAISTKR
PARKKAIST / SAMSUNG ELECTRONICS KR
YUNG-LYUL LEEKAIST / SAMSUNG ELECTRONICS KR
CHU-KEON CHEONGLG SIMICONDUTOR KR
HYUN MUN KIMLG SIMICONDUTOR KR
JEON BYEUNGWOOSAMSUNG ELETRONICS KR
JANG EUEE SEONSAMSUNG AIT KR
JAE-SEOB SHINSAMSUNG AIT KR
SANG-WOOK KIMSAMSUNG AIT KR
YEON-BAE KIMSAMSUNG AIT KR
YOUNG-RYUL LEESAMSUNG AIT KR
ROB KOENENKPN RESEARCH NL
ROB A. BEUKERPHILIPS NL
RAKESHI TAORIPHILIPS RESEARCH NL
WERNER OOMENPHILIPS RESEARCH NL
LEON VAN DE KERKHOFPHILIPS S&V NL
ASMUND SANDVANDTELENOR NO
ROBERT DANIELSENTELENOR R.D NO
FERNANDO PEREIRAINSTITUTO SUPERIOR TÉCNICO PT
PAULO NUNESINSTITUTO SUPERIOR TÉCNICO PT
MORGAN LINDQVISTERICSSON RADIO SYSTEMS SE
BO BURMANERICSSON RADIO SYSTEMS AB SE
GORAN ROTHERICSSON RADIO SYSTEMS AB SE
HARALD BRUSEWITZERICSSON RADIO SYSTEMS AB SE
PER THORELLERICSSON TELECOM AB SE
TORBJÖRN EINARSSONERICSSON TELECOM AB SE
KLAS HYLLANDERTELIA RESEARCH AB SE
MARIE WILHELMSSONTELIA RESEARCH AB SE
OLLE FRANCESCHITERACOM SVENSK RUNDRADIO SE
SOFIE OLSSONTERACOM SVENSK RUNDRADIO SE
TAN THOW KENGPANASONIC SINGAPORE LABORATORIES SG
PATRICK MULROYBT LABORATORIES UK
DAVID MEARESBRITISH BROADCASTING CORPORATION UK
MIKE KEITHINTEL CORPORATION UK
SIMON WINDERPACTUK
PAUL FELLOWSSGS - THOMSOM MICROELETRONICS UK
MIROSLAW BOBERUNIVERSITY OF SURREY UK
ATUL PURIAT&T LABS USA
JOERN OSTERMANNAT&T LABS USA
S. QUACKENBUSHAT&T LABS USA
BARRY G. HASKELLAT&T LABS USA
JAMES D. JOHNSTONAT&T LABS USA
MUKTA KARCABLE TELEVISION LABORATORIES INC. USA
ALEXANDROS ELEFTHERIADISCOLUMBIA UNIVERSITY USA
IRAJ SODOGARDAVID SARNOFF RESEARCH CENTER USA
YA-QIN ZHANGDAVID SARNOFF RESEARCH CENTER USA
JOHN D'ALICANDRODELTA INFORMATION SYSTEMS USA
RICHARD SCHAPHORSTDELTA INFORMATION SYSTEMS, INC USA
FREDERIC DUFAUXDIGITAL EQUIPAMENT CORP USA
CHRISTOPHER ADAMSDIVICOM USA
MATTHEW GOLDMANDIVICOM USA
MARINA BOSIDOLBY LABORATORIES USA
ROBERT SENNEASTMAN KODAK COMPANY USA
PETER K. DOENGESEVANS & SUTHERLAND USA
MIKE COLEMANFIVE BATS RESEARCH USA
AJAY LUTHRAGENERAL INSTRUMENT USA
MANDAYAM NARASIMHANGENERAL INSTRUMENT USA
XUEMIN CHENGENERAL INSTRUMENT USA
JEFF HAMILTONGENERAL INSTRUMENT USA
GANESH RAJANGENERAL INSTRUMENT USA
DONALD MEADHUGHES ELECTRONICS USA
RONNIE BURNSHUGHES ELECTRONICS USA
JIM C. WILLIAMSHUGHES ELETRONICS USA
PETER H. AUHUGHES ELETRONICS USA
RAM NAGARAJANHUGHES ELETRONICS USA
JEFFREY LYNCHIBMUSA
PETER P. SCHIRLINGIBM USA
CHRIS HANSENINTEL CORPORATION USA
THOMAS GARDOSINTEL CORPORATION USA
MICHAEL ZEUGITERATED SYSTEMS USA
JOHN MULLERITERATED SYSTEMS USA
WEIPING LILEHIGH UNIVERSITY USA
ERIC PETAJANLUCENT TECHNOLOGIES USA
CASPAR HORNEMEDIAMATICS, INC USA
MING-CHIEN LEEMICROSOFT CORP. USA
WEI-GE CHENMICROSOFT CORP. USA
DAVID THOMMITSUBISHI USA
ROBERT BELLMITSUBISHI USA
HUIFANG SUNMITSUBISHI ELECTRIC, ITA USA
JIM BRAILEANMOTOROLA USA
CHEUNG AUYEUNGMOTOROLA USA
KEVIN O'CONNELLMOTOROLA USA
MARK BANHAMMOTOROLA USA
OTTO SCHNURRMOTOROLA USA
JIRO KATTONEC CORP. / PRINCETON UNIVERSITY USA
HOMER CHENROCKWELLUSA
TIMOTHY ADDINGTONSCIENTIFIC ATLANTA USA
PETER SCHREINERSCIENTIFIC ATLANTA USA
IBRAHIM SEZANSHARP LABS OF AMERICA USA
AMAN JABBISUN MICROSYSTEMS USA
JAMES VAN LOOSUN MICROSYSTEMS USA
RADU JASINSCHITEKTRONIX, INC. USA
CHARLES LUECKTEXAS INSTRUMENTS USA
FRITZ WHITTINGTONTEXAS INSTRUMENTS USA
RAJ TALLURITEXAS INSTRUMENTS USA
IOLE MOCCAGATTATEXAS INSTRUMENTS, INC USA
OSAMA ALSHAYKHUNIVERSITY OF CALIFORNIA, BERKELEY USA
RALPH NEFFUNIVERSITY OF CALIFORNIA, BERKELEY USA
AHMET MURAT TEKALPUNIVERSITY OF ROCHESTER USA


Annex 2

Meeting agenda
1. Opening
2. Roll call of participants
3. Approval of agenda
4. Allocation of contributions
5. Communications from Convenor
6. Report of previous meeting
7. Processing of NB Position Papers
8. MPEG Phase 2
8.1Audio
8.2Verification of MPEG-2
8.2.1 Video Quality
8.2.2Audio Quality
8.3Amendments
8.3.1Private Data (System #3)
8.4Part 6 (DSM-CC)
8.5Part 7 (NBC Audio)
8.6Part 10 (DSM-CC Conformance)
8.7Workplan
9.MPEG Phase 4
9.1Requirements
9.2Tools
9.2.1Systems
9.2.2Natural Audio
9.2.3Synthetic Audio
9.2.4Natural Video
9.2.5Synthetic Video
9.2.6Joint matters
9.3Verification Models
9.3.1System
9.3.2Video
9.3.3Audio
9.3.4SNHC
9.4Tests
9.5Call for proposals
9.6Simulation software
9.7Working Draft
9.7.1System
9.7.2Video
9.7.3Audio
9.7.4SNHC
9.8Workplan
10.MPEG Phase 7
11. Overall WG11 workplan
12. Liaison matters
13. Administrative matters
13.1 Schedule of future MPEG meetings
13.2Promotion of MPEG
14. Organisation of this meeting
14.1 Tasks for subgroups
14.2 Finalisation of meeting allocation
15. Planning of future activities
16. Resolutions of this meeting
17. A.O.B
18. Closing


Annex 3

List of submissions
No.
Source
Title
1369Pete SchirlingContribution Register for 37th Meeting in Maceio, Brazil
1370Ming-Chieh Lee, Wei-ge Chen, Chuang Gu Microsoft Software C++ Implementation of MPEG-4 Video VM4.x
1371David Kirby (BBC), Kaoru Watanabe (NHK) Report on the Formal Subjective Listening Test of MPEG-2 NBC multichannel audio coding
1372Hendrik FuchsReport on Investigations on Predictor Complexity Reduction
1373Jean-Claude Dufourd MSDL-S Frequently Asked Questions
1374P. G. Schreiner III, H. Suzuki Report of the Ad Hoc Group on MPEG-2 NBC Formal Listening Test
1375AFNOR - FRANCEFrench National Body proposition for the handling of copyrighted works i
1376Fritz Seytter2-Layer Multiplex for MPEG-4
1377Jean-Claude Dufourd (France Tilicom/ENST), Gilles Privat (France Tilicom/CNET), on behalf of Emphasis MPEG4 terminal architecture
1378Marina BosiReport of the Ad-hoc Group on MPEG-2 Audio NBC (13818-7) Committee Draft Progression
1379Marina BosiReport of the Ad-hoc Group on MPEG-2 Audio NBC (13818-7) Committee Draft Progression
1380Andre Kaup, Anke Lorenz Results of Core Experiment O9
1381Vahe BalabanianReport on DSM-CC-2 Ad Hoc
1382Mark Banham, James Brailean Results for Error Resilient Core Experiment E6
1383Werner Oomen, Rene van der Vleuten, Leon van de Kerkhof Results of the MPEG2-NBC Core experiment on adaptive Huffman coding
1384John Villasenor, Brendan Dowling, Max Luttrell(UCLA), Dong-Seek Park(Samsung Elec. Co.) Simulation Results for H.223/Annex A
1385John Villasenor, Judy Liao Use of H.223 over wireless links: Proposal for H.324/Annex C Modes 2 and 3
1386TK Tan, SM ShenResults of Core Experiment N1
1387TK Tan, SM ShenResults of CE B1.1 and Merged Mode for Spatial Scalability
1388TK Tan, SM ShenMerged Mode in Spatial Scalability
1389Shigeki FujiiMusic Synthesis for MPEG4 SNHC
1390G.Russo, S.Colonnese, U.Mascia, C.Tabacco Updated FUB Results on Core Experiment N2 on Automatic Segmentation Techniques
1391Shi-Hwa Lee, Euee-Seon Jang, Yu-Shin Cho, Se-Hoon Son,, Dae-Sung Cho, Jae-Seob Shin, Yang-Seock Seo Results of CE S4 : Comparison of Shape Coding Techniques
1392Sung-Gul Ryoo, Jae-Seob Shin, Yang-Seock Seo Results of CE Q2 : Improved Rate Control
1393Yung-Lyul Lee, Hyun-Wook Park, Hee-Yong Lim, Yoon-Soo Kim Results of Core Experiment N1
1394Seong-Jin Kim, Euee-Seon Jang, Yang-Seock Seo Results of CE T12 : MC Error Suppression Technique
1395John MullerReport of the AdHoc group on core experiments on efficient coding in MPEG-4 video
1396Karlheinz Brandenburg Report of Ad-Hoc Group on core experiments for MPEG-4 audio
1397Vahe BalabanianImplication of full integration of DSM-CC with ATM in DSM-CC-2
1398Vahe BalabanianITU-T IN Interworking with DSM-CC-2
1399Vahe BalabanianMultimedia Conferencing in DSM-CC-2
1400Vahe BalabanianCurrent Status of ITU-T T.130 Real-Time Audio-Visual Control for Multimedia Conferencing - Architecture & Overview
1401Vahe BalabanianLiaison on the definition of DSM-CC PassThru type values for T.120
1402Richard J. Qian, M. Ibrahim Sezan Scalable Vertex-Based Shape Coding
1403James BraileanReport of ad-hoc group on core experiments on error resilience aspects in MPEG-4 video
1404Regis J. Crinon, Dean Messing, Ibrahim Sezan Results of N3 Experiment using Unified S2/N3 Syntax
1405Koji Imura(1), Yutaka Machida(1), Toshiaki Watanabe(2), Yoshihiro Kikuchi(2), Toshio Miki(3), Tomoyuki Ohya(3), Satoru Adachi(3) The Framework of the Core Experiment(E3b)
1406Jean-Francois VIAL Fine granularity scalability: proposal for a new tool
1407Aasmund Sandvand Results of Core Experiment N1
1408Masayuki Nishiguchi, Kazuyuki Iijima, Jun Matsumoto Use of "Mixed Voiced" mode for 2.0Kbps parametric core of the VM
1409Torbjorn Einarsson, Per Thorell, Carl-Johan Westelius Improved Video Object Syntax
1410Torbjorn Einarsson, Per Thorell, Carl-Johan Westelius Results from CE B1.1 on Spatial Scalability
1411Toshiaki Watanabe, Yoshihiro Kikuchi Comparison of Binary Shape Coding (Core Experiment S4d)
1412Toshiaki Watanabe,, Yoshihiro Kikuchi Comparison of Error Resilient VLC (Core Experiment E3a)
1413Rob Beuker, Ton Kalker, Rene Klein Gunnewiek, Rik Theunis Intra-frame and inter-frame results on core exp. T8: variable-size lapped transform coding
1414Toshiaki Watanabe,, Yoshihiro Kikuchi Efficiency of Error Resilient Resynchronization Code (Core Experiment E6)
1415Koji Imura,Yutaka Machida The Results of the Core Experiment E6
1416Jean-Claude Dufourd, on behalf of Emphasis A new type of synthetic content: short note on Flexes
1417Jean-Claude Dufourd, on behalf of Emphasis A tentative bitstream access API in Java
1418Jean-Claude Dufourd, on behalf of Emphasis Missing concepts in the current MSDL Architecture
1419Noel BradyResults of CE S4f: Context-based Arithmetic Encoding
1420M. SchumannInformation about DSM-CC Conformance Testing activities (DAVIC)
1421A. KnollTiming considerations
1422Xuemin ChenA brief report on core experiment Q2 --improved rate control
1423Ram NagarajanResults of Core Experiment S4i: Grayscale Shape Coding Effects
1424David ThomReport of the ad hoc group on NBC Core Experiment on Huffman Coding
1425Eishi Morimatsu, Akira Nakagawa Dynamic Resolution Conversion Scheme for Low Latency Video Coding
1426Teruhiko Suzuki, Yoichi Yagasaki The results of CE B1.1 and proposal for object based scalability
1427Teruhiko Suzuki, Yoichi Yagasaki The syntax modifications for efficient coding of VOP informations
1428Hirohisa Jozawa, Kazuto Kamikura Results of Core Experiment P1
1429Keiichi HIBI, Nobuyuki EMA Report of core experiment on P6 "Multimode Warping Prediction"
1430Shinya Kadono, C.S.Boon Results on Shape Core Experiment S4d (Improved MMR Methods)
1431Shinichi SAKAIDA, Wentao ZHENG, Yutaka KANEKO, Yoshiaki SHISHIKUI Results of Core Experiments O12 and O13
1432Peter ListProposal for a revised B-Vop option
1433Minoru Etoh and Takeshi Ankei Results of Core Experiment S4i:Grayscale Shape Coding Effects
1434Kazuhisa Hosaka, Yoichi Yagasaki Results of CE S4f (CAE)
1435Naoya Tanaka, Koji Yoshida Proposal for the Speed Control Functionality on the MPEG-4 Audio Core Experiment
1436Jens SpilleMPEG 2 Layer II bitstream verification for NBC formal listening test
1437Jens SpilleHuman Sound Localization
1438Laura ContinReport of the ad hoc group on MPEG-4 July '97 audio visual tests
1439M. Zhou, J. De Lameillieure, T. Sikora HHI-Results on Core Experiment P9
1440Laura ContinTest methods and procedures for July '97 MPEG-4 tests
1441Julien Signes, CCETT, Jiro Katto, Nec Corporation, Neil Day (RICOH) Report of the Adhoc Group on MPEG-4 architecture evolution
1442Gerhard StollReport of the Ad-Hoc Group on 13818-3.2 DIS Progression
1443DIN NI29Resolutions of the German National Body (DIN)
1444Dominique CURETMPEG4 requirements and the MPEG2 transport syntax
1445Marco Mattavelli, Sylvain Brunetton Controlling decoding power by graceful degradation techniques
1446Anthony Vetro, Huifang Sun, Jay Bao Results of Core Experiment Q2: Improved Rate Control
1447Liam Ward on behalf of MoMuSys Donation of MoMuSys Video VM Software to ISO/IEC
1448Bernhard GrillReport of the Ad-hoc Group on MPEG-4 Audio Working Draft Editing and Verification Model Software Implementation
1449Schuyler Quackenbush Report on the Ad hoc group on Reducing NBC Prediction Complexity
1450Michael ZeugReport of Ad Hoc Group on MPEG4 Low Delay Evaluations
1451Donald C. MeadSupercompression Via A Priori Agreed Upon Generic Objects
1452Donald C. MeadDownloading of Personalization Layers
1453Peter AuVideo Compression with Pre-agreed Upon Generic Objects
1454Ronnie R. BurnsResponse to CFP on SNHC Audio
1455Ram NagarajanStandards Adaptive Encoding and Decoding System
1456David Shu, Peter Au Results on Core Experiment S2 Plus - Low Latency Video Coding with Sprites
1457David Shu, Peter Au Results on Core Experiment S2 Plus - Low Latency Video Coding with Sprites
1458Frederic DufauxResults for Video Coding using Dynamic Sprite (Core Experiment N3)
1459Jie LiangStatus Report on Core Experiment "Wavelet Coding of I Pictures" (T1)
1460Raj Talluri, Iole Moccagatta, Yashoda Nag Results on Core Experiment on Error Concealment and Resynchonization Syntax for Separate Motion Texture Mode (E5)
1461Weiping Li, F.Ling,H.Sun,J.P.Wus Report on Core Experiments T5/T6
1462Weiping Li, F.Ling,J.P.Wus Report on Core Experiment T11
1463Donald C. MeadBest Basis Functions as Part of a System for Content Based Storage and Retrieval
1464Caspar Horne, Touradj Ebrahimi Report of the AdHoc Group on SNHC VM 1.0 Specifications Editing
1465Jim Scholl, Wei Wu, Homer Chen Results of Core Experiment T1
1466Wei Wu, Homer Chen Results of N-Bit Video Coding
1467Murat Tekalp, Homer Chen Report of the Ad Hoc Group on Joint SNHC-Video Technical Issues
1468M. Ogata, N. Miyahara, A. Mackinnon Results of core experiment T1
1469Tsuyoshi KASAHARA Results of Core Experiment P2&P3
1470Zhixiong Wu, Toshifumi Kanamaru Results of Core Experiment T13
1471Zhixiong Wu, Toshifumi Kanamaru Results of Core Experiments O12 and O13
1472Itaru Kaneko(itaru-k@ascii.co.jp) Chairman's report of Ad-hoc group on the SNHC Audios
1473Itaru Kaneko(itaru-k@ascii.co.jp) Report on the internal discussion. Publisher's view on the copyrights, fonts, etc.s
1474Portuguese National Body (IPQ) Portuguese National Body Position
1475 Luis Torres, David Gimeno Efficient texture coding using SVQ
1476 Mark Banham, Rick Dean Results for Core Experiment E5: Error Concealment and Resynchronization Syntax for Separate Motion Texture Mode
1477Caspar Horne and Touradj Ebrahimi (editors) Draft Specification of SNHC VM 1.1
1478 Michael R. Frater, John F. Arnold Draft Requirements for Surveillance Profile
1479Michael R. Frater, John F. Arnold Syntax, Semantics and Experimental Conditions for n-Bit Video Coding
1480Michael R. Frater, John F. Arnold, Martin Kuchlmayr Results from Coding of 12-Bit Thermal Imagery
1481Taisuke MATSUMOTO, Shinya Kadono, Minoru Etoh, C.S.Boon Results on Shape Core Experiment S4f (Bitmap-based Method)
1482Jinhun Kim, Sanghoon Lee, Jongil Kim Results of core experiment S4a:Geometrical Representation Method
1483C.S.Boon, J.Takahashi Software for Macroblock-based Padding
1484 C.S.Boon, M.Okuno,S.Kadono Audiovisual Sequences for July 97 Tests
1485C.S.Boon, T.Takahashi Update of Flex_0 Composition Layer Syntax
1486C.S.Boon, J.Takahashi Experiment Results of Core Experiment O2
1487C.S.Boon, J.Takahashi Experiment Results of Core Experiment O9
1488C.S.Boon, M.Okuno Experiment Results of Core Experiment O14
1489C.S.Boon, T.Nishi,S.Kadono Scaling of Quantisation Parameter for Grayscale Alpha Plane : A New Proposal
1490C.S.Boon,J.Takahashi,S.Kadono Padding for Motion Compensation : Review of Past Results and Preliminary Study of Complexity Evaluation
1491Jae-won Chung, Jin-hak Lee, Jae-kyoon Kim, Joo-hee Moon Results of Shape Coding Core Experiment : S4a and S4h
1492Toshio Miki, Toshiro Kawahara, Tomoyuki Ohya Revised Error Pattern Generation Programs for Core Experiments on Error Resilience
1493Yuji Itoh Results of Core Experiment N1
1494 Yuji Itoh Results of Core Experiment S3
1495 Thomas Stingl, Doris Bittruf, Yankin Tanurhan, Alexandra Winter MPEG-4 Interactivity Aspects
1496 MoMuSys' partners MoMuSys C implementation of MPEG-4 video VM 4.0
1497Marta Karczewicz Report on Core Experiment P9
1498 Kinya Oosa A low-complexity deblocking filter
1499 Touradj Ebrahimi (editor) MPEG-4 Video Verification Model Version 4.1
1500Touradj Ebrahimi Report of Ad Hoc Group on Video VM Editing
1501Frederic Jordan,Touradj Ebrahimi Viewer dependent texture coding for transmission of virtual environment
1502Keiichi HIBI, Tadashi UCHIUMI Report of Error Resilience core experiment E6
1503Hiroyuki Katata, Ryuji Kitaura Report of the core experiment on Object-based Temporal Scalability
1504Hiroyuki Katata, Shuichi Watanabe, Toshio Nomura Proposal of the syntax of Spatio-Temporal Scalability
1505Nobuhiko Naka, Toshio Miki A Synchronization Code Implementation for Error Resilience Aspects in MPEG-4 Audio VM
1506Takahiro Fukuhara, Kohtaro Asai, Shun-ichi Sekiguchi, Tokumichi Murakami Results of Core Experiment P2 and P3
1507 Shun-ichi Sekiguchi, Kohtaro Asai, Takahiro Fukuhara, Tokumichi Murakami Results of Core Experiment P11
1508 Takehiro Moriya, Satoshi Miki, Naoki Iwakami, Akio Jin, Kazunaga Ikeda Proposals of core experiments for MPEG-4/audio VM
1509Toshiyuki Nomura, Kazunori Ozawa, Masahiro Serizawa, Masahiro Iwadare Proposal of compression algorithm with rate control for MPEG-4/Audio core experiments
1510Toshio Miki, Satoru Adachi, Tomoyuki Ohya Results of Core Experiments on Error Resilience
1511Peter Kuhn,, Andreas Hutter Results of Core Experiment P9 (Motion Compensated Prediction using Quadtree Segmentation and Polynomial Motion Fields)
1512Ulrich Benzler Results of core experiment P8 (Motion and Aliasing compensating Prediction)
1513C.Stoffers, T.Sikora Results of core experiment P8
1514 Ulrich Benzler Improvement of spatial scalability by subband decomposition
1515Bertrand Klaiber, on behalf of EMPHASIS Thread Primitives in the Java Environment
1516Jae Gark Choi, Young-Kwon Lim, Myoung Ho Lee, Gun Bang, Jisang Yoo Automatic segmentation based on spatio-temporal information
1517Carsten Herpel (THOMSON multimedia),, Philip A. Chou (Xerox PARC) Elementary Stream Naming Convention
1518 Carsten Herpel Report of AHG on MSDL Multiplex Specification and Signaling
1519Jan De Lameillieure HHI results of Shape Adaptive and M*N DCT (O4 and O8)
1520Y. Suzuki, Y. Nakaya, S. Misaka, A. Date Results of Core Experiment P1(Global Motion Compensation)
1521Y. Nakaya, Y. Suzuki, A. Date, S. Misaka Results of core experiment P6 (multimode warping prediction)
1522Miroslaw Bober, Kui Zhang, (on behalf of the SCALAR consortium) Region-based motion compensation using fixed patterns.
1523Rakesh Taori, Andy Gerrits Bit-rate scalability in LPC-ABS wideband coder
1524Minsoo Hahn, Jungchul Lee MPEG4 TTS Interface
1525 Marco Mattavelli,, Stefano Battista,, Franco Gasperoni,, Bertrand Klaiber,, Laurant Pautet,, Samuel Tardieu Concurrent versus Sequential Modeling for MSDL
1526Marco Mattavelli Report of the ad-hoc group on computational graceful degradation
1527Angelika Knoll Some ideas for a simple buffer model
1528 Yoshihiro Miyamoto Results of Core Experiment P6 (Multimode Warping Prediction)
1529Carsten Herpel Multiplex Interface for Systems VM
1530 Carsten Herpel MSDL-M Adaptation Layer Syntax
1531 Carsten Herpel System Decoder Model for Resource Management
1532Carsten Herpel Multiplex-centric View of Data Flows in MPEG-4 System
1533Rob Koenen Report of AHG on MPEG-4 requirements
1534 Harald Brusewitz Results on CE E3d (Robust FLC)
1535 Harald Brusewitz Proposal for OBMC flag
1536 Gunnar Nitsche, Peter Mangold An interleaving tool for MPEG-4 MUX
1537 Rainer Puellen, Detlef Krahe & Partner Audio on Demand, Multimedial Infosystems, Media NRW
1538Sven Bauer, Gunnar Nitsche MPEG-4 Video Profiles and Levels
1539 Iraj Sodagar, Hung Ju Lee, Dinei Florencio, Paul J. Panaro Results of Core Experiment T1: Wavelet Coding of the I Pictures
1540Iraj Sodagar, Dinei Florencio, Hung Ju Lee, Paul J. Panaro Results of Core Experiment T2: Wavelet Coding of the P Pictures
1541Tihao Chiang, Iraj Sodagar, Ya-Qin Zhang Results of Core Experiment Q2: Improved Rate Control
1542Roberto Castagno Results of Core Experiment N1
1543 Christopher Ward Digital Studio Command and Control Proposed Activity
1544Wei-ge Chen, Ming-Chieh Lee Efficient Transparent Block Skipping
1545 Wei-ge Chen, Ming-Chieh Lee, Chuang Gu Insight on Sprite Technology and Relation to Video and SNHC
1546Chuang Gu, Ming-Chieh Lee Results on Core Experiment O2: Padding for Motion Compensation
1547Chuang Gu, Ming-Chieh Lee Results on Core Experiment O14: Original vs. Reconstructed Shape for DCT Padding
1548Stefano Battista & others (MoMuSys) MSDL VM 2.0: Proposals
1549 Peter Gerken, Roland Mech Automatic segmentation of moving objects (Partial results of core experiment N2)
1550Bernhard Grill Update of the preleminary MPEG-4 Working Draft
1551Ram Nagarajan Report of AHG on Media Integration of Text and Graphics
1552A. Hutter, C.S. Boon (Matsushita) MPEG4 Video Format Conversion Software V1.3
1553A. Eleftheriadis, Y.Fang A Proposed Revision of MSDL-S
1554 Andreas Hutter, Peter Kuhn, Stefan Herrmann Some experiments for CE P5
1555 A. Eleftheriadis The Benefits of Using MSDL-S for Syntax Description
1556A. Eleftheriadis, Y. Fang The Integration of MSDL-S with MSDL Architecture and VM
1557A. Kopansky, K. Wu, J.-B. Lee, Y. Fang, A. Eleftheriadis The MPEG-4 Video VM 4.0 Syntax Using MSDL-S
1558Peter Gerken, Hailian Li Comparison of shape coding techniques (Partial results of core experiment S4)
1559Lin Yin, Mauri Vaananen Proposal for a Core Experiment on Backward Prediction in MPEG-4 Audio
1560Peter K. Doenges, Ganesh Rajan Report of the Ad Hoc Group on Synthetic/Natural Hybrid Coding
1561Joern Ostermann Report of the adhoc group on core experiments on object- or region-oriented texture coding in MPEG-4 video
1562Joern Ostermann Report of the adhoc group on core experiments on shape and alpha channel coding in MPEG-4 video
1563A. Puri Report of Ad hoc Group on Multifunctional Coding in MPEG-4 Video
1564A. Puri Report of Ad hoc Group on Video Working Draft Editing
1565A. Puri, R. L. Schmidt, B. G. Haskell Results and description of Experiment T11
1566Frank Bossen Improved block-based CAE shape coding: don't flush that pipe!
1567P. J. L. van Beek, A. M. Tekalp Revised syntax and results for CE M2 (Triangular mesh-based coding)
1568Ralph Neff, Osama Al-Shaykh, Eugene Miloslavsky, Emin Martinian, Shang-Pin Chang, Avideh Zakhor Core Experiment T3: Current Status and Results
1569Hai Tao, Thomas S. Huang, Homer Chen, Wei Wu, Tsae-Pyng Janice Shen Experiments of FAPs coding
1570 G. H. Park Core Experiment Results of T12: MC Error Suppression
1571Jan De Lameillieure Results of adaptive scan orientation in MMR shape coding (S4d)
1572Kevin O'Connell Results for S4a - Vertex-Based Shape Coding
1573Secretariat, ISO/IEC JTC 1/SC 29 Summary of Voting on ISO/IEC 13818-2/PDAM 4, Extension code assistant
1574Secretariat, ISO/IEC JTC 1/SC 29 Summary of Voting on ISO/IEC CD 13818-7, Extension for non backward compatible audio coding
1575Edward Cooke,, Stephane Lecercle,, Olivier Avaro,, on behalf of MoMuSys MSDL VM : A Primer
1576 Edward Cooke,, Stephane Lecercle,, Olivier Avaro,, on behalf of MoMuSys MSDL VM : A Primer
1577 Cheung Auyeung Report of the statistic and bitstreamm exchange ad hoc group
1578Simon A.J. Winder Report of the Ad-Hoc Group on Investigating Reduced Complexity Padding Techniques
1579Olivier Avaro Report of the AHG on MSDL Working Draft Editing
1580Olivier Avaro Report of the AHG on MSDL Verification Model
1581G. Bjontegaard Improved INTER VLC coefficient coding with automatic switch between two VLC's
1582G. Bjontegaard Proposal for new INTRA VLC
1583 Haibo Li,, Franck Davoine,, Robert Forchheimer Facial Texture Coding
1584 Richard Schaphorst H.263+ Liaison to MPEG-4 Video
1585 Richard Schaphorst H.263L Input to MPEG-4 Requirements
1586 Richard Schaphorst Liaison to MPEG-4 re Multiplex
1587 Richard Schaphorst Liaison to MPEG on MPEG-4 Systems Issues
1588Richard Schaphorst Liaison to MPEG on Description of MPEG-4 (N1410)
1589John Muller H.263+ liaison to MPEG-4 video
1590 Tsuhan ChenLBC Documents on Chroma Keying


Annex 4

Report of Requirements meeting

Source: Rob Koenen, Chairman MPEG Requirements

1. General

Again more people found time to spend at least a part of the meeting with the requirements group. This is considered to be a good development, because more and more decisions will be taken in this group, and this is where delegates can explain and represent their and their companies' interests.

The main topics of the meeting were:

It became apparent that the role of the requirements group is not yet entirely clear to all in MPEG. The requirements group decided to take in consideration all contributions directed to the group this time - even when they were brought to the attention of the group only during the meeting - but the chairman will be more selective at the next meeting.

2. General Requirements Issues & Requirements Document

Much work was carried out by the SNHC group to get the SNHC Requirements in the General Requirements Document in good shape. A vivid discussion resulted in the reassurance by the SNHC people that they could deliver all the technology that is currently in the Requirements Document.

Another lively discussion, with the Audio Group, resulted in their promise to fill in the numerous 'to be decided's in the General Requirements, before the meeting in Seville.

A new version of the requirements document was issued (N1495). An ad hoc group was created to work on requirements and profiles, chaired by Rob Koenen (N1496)

The requirements group recognised that the proposal covered in contributions M1451 and M1452 (pre-defined objects together with personalisation layers) addresses the requirement for having very low bitrate speech and perhaps audio coding. It recommends that the Audio and SNHC evaluate the technology, that should be described in more detail, as currently only a very generic description of the technology is given.

Sprite technology was discussed (contribution m1545) and the requirements group recommended that the Video Group consider sprite technology as a tool for representing pixel-based Video. The requirements group sees usage in the context of Real-Time Communications (in which coding efficiency is a prime requirement) and multimedia broadcast.

In a joint meeting with Systems and DSM-CC it was decided to initiate an ad hoc group to investigate signalling requirements for MPEG-4 (N1490), chaired by Carsten Herpel and James Van Loo.

In joint meetings with Systems, the issue 'flexibility' was discussed several times. The flex_levels have disappeared, and the issue of flexibility needs further study, as there is no consensus yet. The danger is that in introducing flexibility, it will be possible to download, following the specifications of the standard, complete decoders in machine-independent code, the performance of which can impossibly be guaranteed, while it is unclear when hardware will be powerful enough for such a requirement to be useful. On the other hand, the clear wish exists to download, in a standardised and machine-independent manner, object behaviour to the decoding machine.

As low latency issues were considered of general interest to MPEG-4, extending beyond the Video Group, an ad hoc on 'low delay evaluations' was set up by the Requirements Group, chaired by Mike Zeug (N1497).

The input from ITU-T SG 15 LBC was welcomed, and input for a responding liaison was written. The suggestions with regard to the Real-Time Communications profile have been incorporated in the profile document. The suggestion to standardise different video coders for different applications and bitrates was not completely followed; the requirements group is of the opinion that as many applications and bitrates as possible should be supported, but that it is desirable to do so with the minimum amount of different algorithms.

3. MPEG-4 Profile Development

Most of the time in the requirements group was dedicated to discussions about profiles, both in the requirements meetings and in subgroups.

Also, the general profile policy was discussed. Profiles collect a subset of requirements that should be met at the same time. These subsets are relevant for one or several classes of applications. The objective is achieving interworking, and profiles thus specify a conformance point.

The important issue in trying to accommodate several classes of functionalities, is whether the requirements for these classes are similar (and not whether the applications themselves are.) In trying to fit an application in with an existing profile, the general policy is to put as many tools in one profile as possible, as long as it does not overly burden applications that do not need the tools.

Currently, there are no profile definitions, but only requirements for profiles, mixed with descriptions of applications. These will have to be translated to real profile descriptions, to be included in the normative parts of the MPEG-4 standard.

New profiles were created following requirements coming from the applications 'broadcast' and 'content-based storage and retrieval'. For the latter, many attributes were identified that give such information about objects as when they were recorded, how, with which camera, etc. I must be decided which of these attributes deserve standardisation by MPEG, and which should be 'private data'.

The broadcast profile is expected to give considerable savings in bitrates for broadcasting of information that contains graphics, (scrolling) text, and some natural video. The savings will come from the ability to separately code and transmit the different types of visual data.

A placeholder was created for the surveillance application that requires 12 bit luminance coding. The requirements group is hesitant about defining a separate profile, and wants to investigate whether its requirements can be met by another, existing profile, notably the storage and retrieval profile.

Contribution m1538

The requirements group decided to define a new level for the "Real-Time Communications profile, level 2, with resolutions going up to CCIR 601 and bitrates up to 2 Mbit.

A new version of the profiles document was issued (N1494).

4. Content-related IPR issues

The requirements group recognised that it is important to provide mechanisms that allow protecting the Intellectual Property Rights attached to MPEG-4 encoded content. Like in MPEG-2, it shall at least be possible to attach a unique identifier to an object, also server a pointer to organisations that represent the interests of content creators.

Several IPR experts were present at the meeting, and working within the requirements group, they drafted an initial assessment of IPR requirements in MPEG-4 (N1428). An Ad Hoc group created on IPR issues was established (N1491), that is chaired by one of the experts: Dominique Yon.

5. Call for Proposals (N1499)

A call for Proposals, to be evaluated in July 97, was written and issued during the meeting. This Call was written after joint meetings with the Test group, and also with Audio and Video. Proposals are welcomed that:

6. MPEG-7

Two very inspiring, fruitful and well-visited sessions were held to discuss 'MPEG-7'. The result of these two meetings was a much better definition and understanding of its goals, which was laid down in the document N1425: "MPEG-7, Context and Objectives". It was recognised that the step from MPEG-4 to MPEG-7 is a rather logical one, and that MPEG-7 builds naturally on the work on MPEG-4. On the other hand, it will be necessary to involve experts not yet participating in MPEG, and expertise not yet present. The next step will be the organisation of an MPEG-7 seminar.

The MPEG-7 work plan now says: WD - 7/99; CD - 3/2000; DIS - 7/2000; IS - 11/2000.

7. Initial WD Assessment

Together with the standard-producing subgroups (Systems, Audio and Video), an assessment was made of the compliance of the Working Draft to the requirements. Although also SNHC produces standards (to be included with the Audio and Video parts respectively), no assessment of the state of their work was made yet. This will be done at future meetings however. The document reporting on this status is N1492.

8. Documents processed
Requirements and Profiles
1451 Donald C. Mead - Supercompression Via A Priori Agreed Upon Generic Objects
1452 Donald C. Mead - Downloading of Personalization Layers
1453 Peter Au - Video Compression with Pre-agreed Upon Generic Objects
1455 Ram Nagarajan - Standards Adaptive Encoding and Decoding System
1463 Donald C. Mead - Best Basis Functions as Part of a System for Content Based Storage and Retrieval
1538 Sven Bauer, Gunnar Nitsche - MPEG-4 Video Profiles and Levels
1545 Wei-ge Chen, Ming-Chieh Lee, Chuang Gu - Insight on Sprite Technology and Relation to Video and SNHC
1478 Michael R. Frater, John F. Arnold - Draft Requirements for Surveillance Profile
1450 input from low delay AHG
1585 + 1588 ITU LBC liaisons
1441 ... Systems Profiles
Intellectual Property Rights Issues
1375 Afnor France - French National Body proposition for the handling of Intellectual Property Rightsed works ...


Annex 5

Report of Systems meeting

Source: Olivier Avaro

1. Architecture

The architecture of the Systems was improved in the two following areas :

The flex level (0-2) has been suppressed and replaced by fixed and flexible profiles, which we should rather see as the support or not support of tools like a Virtual Machine or a VRML browser. The scope of flexibility has been restricted to the areas where there is market need and technologies foreseen, that is, composition. Other areas of work will be addressed in due time if needed.

The need of 3D composition information VRML like, first analyzed within the architecture evolution group has received a full support and should be one a the major area of work in Systems. Future work on this issue will be carried on in the MAG.

2. Composition

2D fixed composition is now explicitly addressed in the working draft. These specifications should be implemented and validated in the next months. The flexible 2D composition is now completely specified in the Verification Model document.

3. Syntax decoding

The specification of the syntactic description language has been released. Support to the other groups in order to describe their syntax specification have been provided during the meeting and should continue until next meeting in order to rich the full description of the working draft.

4. Multiplex, synchronization and signaling

A quantum leap was made here in the specification that now give a good picture of what we want to achieve. The gap between audio-visual objects and elementary streams has been filled, a timing model has been proposed and the multiplex specification has been enhanced with a two layer model. I am quite confident that now the specification can serve an implementation. Carsten will give here a more detailed report ;-)

Then we have the two documents we wanted to reach : the working draft and the verification model. Congratulations to all of you and many thanks to the editor chairs, Alex, Carsten, Julien, Phil and Stefano and the "cross checkers", Gilles and Jean-Claude.

5. Further activities

The AHG has been set up, with their associated mandates.


Annex 6

Report of Video Meeting

Source: Thomas Sikora, Chairman MPEG Video Group

The MPEG Video Group met in Brazil from 18th - 22nd November. On the 17th November a half day joint meeting was held with the MPEG SNHC Group.

In the joint Video/SNHC Group meeting efforts were synchronize between the two groups and Video Group work items related to SNHC were identified. Based on various presentations from both SNHC and Video and following detailed discussions, the two group identified the need to assist and develop technology for coding of high quality/high resolution still images with provisions for scalability over various levels of quality/bitrate and complexity. A set of Core Experiments was established to investigate suitable technology. Results of the Core Experiments will be reviewed in the next Video Group meetings to ensure that an integrated set of still image coding tools is being developed.

The main topics in the MPEG Video Group meeting were related to the establishment of Version 5.0 of the Video Verification Model (VM) (Doc. 1469) and the establishment of the first version of the MPEG-4 Video Working Draft document.

1. MPEG-4 Video Working Draft

The first Version of the MPEG-4 Video Working Draft (WD) was issued during the meeting and approved by the Video Group (Doc. 1470). The Working Draft is a first draft of the MPEG-4 Video Coding standard detailing the decoder and the bitstream semantics.

2. MPEG-4 Video Verification Model

The following techniques were adopted or altered for the new version 5.0 of the VM:

Improved coding efficiency (rectangular VOP sequences):

Improved coding efficiency (arbitrarily-shaped VOP sequences):

3. Core Experiments

Results of Core Experiments were reviewed. The number of core experiments in the Video Group was reduced substantially to foster the collaborative effort in the group towards few techniques. The remaining list of Core Experiments is detailed in Doc. 1469.

4. July 1997 Tests

A number of new test sequences for the July 1997 Subjective Tests were identified in a joint meeting with the Test group.

5. Reference Software

The VM software implementations donated by MOMUSYS (C version) and Microsoft (C++) have been adopted as Video VM software reference implementations.

6. Future Activities

With respect to discussions in the Requirements group and the outcome of a joint Video/Requirement meeting the Video Group adopted for its future workplan the extension of the techniques under investigation towards substantially higher pixel numbers and bit rates bitrates (up to 20 Mbits/s) and the provisions for coding interlaced input material using suitable inter-field prediction modes.

It is further targeted to develop "generic" scalable still image coding techniques that provide good image quality over a wide range of bit rates suitable for diverse applications, in particular for coding pixel based texture information to assist SNHC functionalities. A close liaison with the JPEG 2000 standardization phase is envisioned.


Annex 7

Report of Audio Meeting

Source:

P. Schreiner, Chairman, Audio Subgroup
D. Meares, Secretary, Audio Subgroup

1. Opening of the meeting

The MPEG/Audio Subgroup meeting was held during the 37th meeting of WG11 in Maceio, Brazil on November 18 to 22, 1996. The list of participants is given in xxxxx. The Chairman welcomed the delegates to the meeting and outlined the work for the five days. He explained the purpose and scheduling of the joint activities with other groups.

2. Approval of agenda

The agenda as presented in xxxxx was discussed and approved after minor changes.

3. Tampere and Chicago meeting reports

The Audio Subgroup portion of the Tampere meeting report, July 1996, had been previously distributed and was approved. The Audio Subgroup portion of the Chicago meeting report, October 1996 was also approved.

4. Allocation of contributions

All contributions were listed (see section xxxxx) and allocated to the agenda. All contributions were presented either in Audio plenary or in the task group discussions.

5. Communications from the Chair

The Chairman summarised the detailed allocations and questions raised at the Chairmen's meeting held on the evening before the main meeting started. The majority of these, by design, were already in the agenda.

6. Ad hoc group reports

All Ad-hoc Group reports (M1374, M1378, M1396, M1424, M1442, M1448, M1449, M1472) had been presented in the opening Plenary. Each was briefly presented again in the Audio Subgroup. Additional details were taken into account in the main part of the meeting.

7. Temporary Task Group Formation

To accomplish the large number of tasks to be performed by the Audio Subgroup, task groups were formed as indicated in xxxxx. The results of each of the task groups were presented to and discussed by the entire audio subgroup, including iterations as necessary. The conclusions of the task groups are presented elsewhere in this report and are included in the output documents.

8. MPEG-2 BC coding

8.1 DIS 13818-3.2

Mr. Stoll reported additional comments on the DIS. The matter raised has been accommodated and is reported in document WG11/N1434, which was approved.

8.2 Copyright release of test bitstreams

Mr. Stoll examined the list of bitstreams used in the compliance tests, to ensure that copyright release had been obtained. During the meeting the remaining bitstreams were cleared of copyright and a list has been released as WG11/N1482. SC29 Secretariat was requested in the final Plenary to make these bitstreams available on CD ROM.

9. MPEG-2 NBC

9.1 Test results

Mr. Meares presented to a joint meeting of Audio and Test the results of the first formal tests of NBC multichannel codecs. This is given in document M1371. There was extensive discussion of some of the details in the report and some requests for changes prior to the approval of the output report. Mr. Meares also reported the additional changes proposed from the BBC during further editorial checks; these were approved. The final version of the report was approved as document WG11/N1419. Test Subgroup reviewed the test report and reported their satisfaction with the work and the report. At the Convenor's request, a brief overview version of these test results was prepared jointly by Test and Audio Subgroups in order to provide material for the Web pages. This was prepared by the task Group shown in xxxxx. Major contributions were made by Messrs Meares, Kirby (by Email and fax from London). This was discussed by the Audio Subgroup and approved as document WG11/N1420. Mr. Spille reported his bitstream verification tests for the NBC tests, M1436. The worst case error was one LSB, verifying the bitstreams. Mr. Meares also reported informal assessment of the stereo derived by down-mixing of the decoded five channels of audio. This test, based on a limited subset of the critical source material, did not show any significant difference to the audibility of artefacts in the five channel auditions. This proved to be very timely as there had been a comment on the CD from the Australian National Body. Mr. Meares drafted a brief summary of this assessment to be included in the Disposition of Comments.

9.2 Options for future developments of NBC

Mr. Oomen presented his document, M1383, on a core experiment on adaptive Huffman coding. The results of the test are positive, but show a benefit in terms of coding gain of between 0.02% to 0.17%, for a cost in terms of encoder processing time of between 1.8% and 11.6%. In general discussion it was noted that the currently achievable benefits for adaptive Huffman coding was extremely small, whilst coding gain for the NEC lossless coding appeared to be less than 0.3% for the sequences tested. Mr. Schreiner therefore asked whether the additional complexity, even if small, should be imposed on all future users of the NBC standard. In other words, should the Audio Subgroup as a professional body of experts continue the optimisation process? Questions were also raised about whether the benefit of adaptive switching of window shape justified its retention. This led to a heated debate on whether or not the group should continue fine tuning of the standard. The consensus of opinion was that those features which had been retained because of earlier core experiments should be retained regardless of whether the benefit proven at an earlier stage of the RM development could still be demonstrated. Thus, NEC lossless coding and adaptive switching of window shape are retained in NBC, but adaptive Huffman coding will not be included. Adaptive Huffman coding could, however, be a useful tool in MPEG-4 at lower bitrates; it will be brought into the VM evaluations. Mr. Quackenbush summarised Mr. Fuch's paper on complexity reduction relating to the use of prediction, M1372. The work that had been done was to see if there were ways of reducing the word lengths used and the number of computation cycles required . The complexity reduction is of the order of 2:1 for ASIC embodiments. The gain for integer machines is less significant.

9.3 DIS preparation

Progress on the DIS continued throughout the week and was presented by Ms. Bosi and her Task Group members to the Audio Subgroup. All of the National Body comments, 288 in total, have been addressed and dealt with in the Disposition of Comments, WG11/N1431. All but 3 comments were accepted. The responses to the comments were then incorporated into the DIS, WG11/N1430. This was presented and approved by the Subgroup.

9.4 NBC Conformance and Technical Report

The Audio Subgroup decided that the NBC IS will require accompanying Conformance and Technical Report elements as amendments to IS 13818-4 and -5. Initial drafting work was carried out to prepare a workplan for these during the meeting. This is presented in WG11/N1424. The copyright released decoder software from AT&T will provide an initial input to the technical report, and AT&T, FhG, Five Bats, and Sony are committed for additional contributions for the encoder and decoder software. The schedule of work appears in the WG11 resolutions of the 37th MPEG meeting.

9.5 New name for NBC

Now that the future of NBC, as a future IS is clear, it is necessary to rename the proposed Standard without the negative style in the name. It was decided to use the name MPEG-2 Advanced Audio Coding (AAC), and this was Recommended to SC29.

10. MPEG-4

10.1 VM activities

The group discussed the format of the VM; should it be based on software modules or a technical description? The former brings the requirement of copyright release as set at the Chicago meeting. The decision, by the end of the meeting, was to base the VM on software modules and to remove from the ftp site all decoder modules that were not available free of copyright restrictions. If there are any missing modules, they will be required to be provided by the February '97 meeting of MPEG or else alternative code will be invited from other MPEG members. This is recorded in output document WG11/N1421. The approved copyright statement will be mapped into all the retained software modules. By the end of the meeting the full range of modules for MPEG-4 Audio had been offered, compliant to the copyright requirement. The Audio Subgroup has implemented the MPEG-4 audio VM in released software. The controlled version of the VM software will be maintained on the Hannover ftp site. The written description of the techniques for the tools is now contained in the WD, and the VM is only software.

It was noted in reviewing the status of the MPEG-4 audio VM that there have already been significant achievements for several of the tools:



MPEG-4 Audio decoder software has been supplied by the following companies:-

AT&T, Deutsche Telekom, Dolby, FhG, Matsushita, NEC, NTT, Philips, Sony, University of Erlangen, University of Hannover


Mr. Brandenburg led a Task Group to review VM proposals and core experiments. He reviewed the Task group work as follows:-

Document M1435 from Matsushita: Proposal for the Speed Control Functionality on the MPEG-4 Audio Core Experiment

Document M1408 from Sony: Use of "Mixed Voiced" mode for 2.0 kbit/s parametric core of the VM

Document M1523 from Philips: Bit-rate scalability in LPC-ABS wideband coder

Document M1509 from NEC: Proposal of compression algorithm with rate control for MPEG-4/Audio core experiments

Document M1383 from Philips: Adaptive Huffman coding

Document M1505 from NTT DoCoMo: A Synchronisation Code Implementation for Error Resilience Aspects in MPEG-4 Audio VM

Document M1508 from NTT: Proposals of core experiments for MPEG-4/audio VM

10.2 Requirements

Mr. Koenen and colleagues joined Audio Subgroup to review the MPEG-4 Call for Proposals. These relate to the new technologies that are to be evaluated in the July '97 tests. The requirements for admission to the tests were modified and the schedule of activities was discussed and confirmed. There may be a fee for participation in these tests, to defray expenses related to tape preparation, listener expenses and statistical analysis.

Mr. Koenen also raised the matter of the MPEG-4 Requirements. A lot of entries read "tbd" in the input document and a wish was expressed that the relevant numbers should be entered. The agreement was that holding numbers would be entered, subject to review by all members prior to the next meeting.

10.3 SNHC

In response to the Call for Proposals, Mr. Burns presented his document M1454, reporting a software product being developed by MIT. This basically presents audio in the form of objects for presentation or onwards processing. Object manipulation is one of the features provided by MIT's approach. Objects are represented by a combination of midi, orchestration and score data trains. These are supported by sound files in the receiver defining individual notes, which can be downloaded in advance or built into ROM. Commercially it uses MIT's C-Sound system in the decoder.

Mr. Kaneko reported on the response to the SNHC audio CFP from Yamaha, document M1389. This work describes a MIDI based music synthesis framework for MPEG-4 SNHC audio. An evaluation experiment and software contributions are proposed for the system elements of a message distributor, a MIDI sequencer, and a MIDI sound generator (object).

The new functionalities addressed by these two proposals that are complementary will be addressed in the SNHC VM.

Mr. Spille presented M1473 on the spatial perception and processing of sound. Head Related Transfer Functions (HRTF) are available on the network for a KEMAR dummy head, so anyone wishing to conduct experiments can access the files. The processing can be applied to either headphones or loudspeakers for limited sitting positions as might be appropriate for computer applications.

10.4

July '97 tests

In a joint meeting, Ms. Contin outlined the preparations for and logistics of the proposed July '97 tests. At the Chicago meeting the first description had been drafted. Document M1440 was prepared as an input to the current meeting. The possibility of tests on mono 64 kbit/s was raised. This brings with it the need to carry out triple stimulus hidden reference testing. All other tests will use the ARC method. No test method is needed for error resilience. Requests were made for new test material. More speech with background noise and music would be beneficial. Mr. Stoll will see if the new EBU material is available. Mr. Masse Suzuki will also check what he can make available. More music items are needed and Mr. Brandenburg suggested that the EBU SQAM disc could be used. 'Pre-segmented' material for object orientated coding will be necessary. Should multichannel be included? If this includes stereo, then the answer is 'yes.' All new material should be brought to the next meeting. Ms. Contin proposed that the matter should be stimulated by issuing a Call for New Test Material, as presented in WG11/N1424. Error conditions need to be discussed. Video are including packet loss, see M1179, and this should also be evaluated for sound. New error patterns should be included and perhaps aligned with Video. Additional features were raised and discussed. These all need to be confirmed at the next meeting. The number of subjects was discussed, with a preference being expressed for 40 listeners.

11. Availability of multichannel test material

A request has been received from a member in Panasonic to seek access to the multichannel test sequences for further MPEG Audio developments and studies. The Subgroup agreed that the copyright owners would have to be contacted before this could happen and Mr. Thom volunteered to add this to the mandate of the NBC Conformance/Technical Report Ad-hoc Group.

12. Consideration of National Body Comments

The NB comments relating to DIS 13818-7 were considered and have been addressed in the DoC. The wish of the Portuguese NB to keep to a minimum the number of ad-hoc groups has been accepted by the Audio Subgroup and, wherever possible, groups were amalgamated.

13. Liaison with other bodies

13.1 ITU-R SG 11, ITU-R WG 10C and DAVIC

A liaison statement was received from ITU-R WG 10C requesting copies of MPEG's NBC and low sampling frequency test results. Liaison statements were prepared for ITU-R SG11, ITU-R WG 10C and DAVIC drawing their attention to the report of the NBC audio codec tests. These are presented in documents N1457, N1455 and N1456. The liaison to ITU-R WG 10C also covers their interest in low sampling frequency codec tests conducted by the Audio Subgroup in 1994. The liaison letter to DAVIC also raises to their attention relevant information for their CFP6.

14. Joint Subgroup Activities

Joint meetings and collaborations were conducted with Test, Implementation, Requirements and SNHC.

15. Press statement

Contributions to the meeting press statement were prepared according to the guidelines in M0883 and approved by the Audio Subgroup.

16. Web page support

A Task Group on FAQs met during the meeting and started to prepare MPEG home page contributions. An initial draft was prepared in WG11/N1433. This work will continue in an ad-hoc group.

17. Contributed documents

The following documents were contributed to the Audio Subgroup:
NumberSource Title
M1371David Kirby (BBC), Kaoru Watanabe (NHK) Report on the Formal Subjective Listening Test of MPEG-2 NBC multichannel audio coding
M1372Hendrik Fuchs Report on Investigations on Predictor Complexity Reduction
M1374P. G. Schreiner III,

H. Suzuki

Report of the Ad Hoc Group on MPEG-2 NBC Formal Listening Test
M1378Marina Bosi Report of the Ad-hoc Group on MPEG-2 Audio NBC (13818-7) Committee Draft Progression
M1383Werner Oomen,

Rene van der Vleuten, Leon van de Kerkhof

Results of the MPEG2-NBC Core experiment on adaptive Huffman coding
M1396Karlheinz Brandenburg Report of Ad-Hoc Group on core experiments for MPEG-4 audio
M1408Masayuki Nishiguchi, Kazuyuki Iijima,

Jun Matsumoto

Use of "Mixed Voiced" mode for 2.0 Kbit/s parametric core of the VM
M1424David Thom Report of the ad hoc group on NBC Core Experiment on Huffman Coding
M1435Naoya Tanaka,

Koji Yoshida

Proposal for the Speed Control Functionality on the MPEG-4 Audio Core Experiment
M1436Jens Spille MPEG 2 Layer II bitstream verification for NBC formal listening test
M1437Jens Spille Human Sound Localisation
M1442Gerhard Stoll Report of the Ad-Hoc Group on 13818-3.2 DIS Progression
M1448Bernhard Grill Report of the Ad-hoc Group on MPEG-4 Audio Working Draft Editing and Verification Model Software Implementation
M1449Schuyler Quackenbush Report on the Ad hoc group on Reducing NBC Prediction Complexity
M1505Nobuhiko Naka,

Toshio Miki

A Synchronisation Code Implementation for Error Resilience Aspects in MPEG-4 Audio VM
M1508Takehiro Moriya,

Satoshi Miki,

Naoki Iwakami,

Akio Jin,

Kazunaga Ikeda

Proposals of core experiments for MPEG-4/audio VM
M1509Toshiyuki Nomura, Kazunori Ozawa, Masahiro Serizawa, Masahiro Iwadare Proposal of compression algorithm with rate control for MPEG-4/Audio core experiments
M1523Rakesh Taori,

Andy Gerrits

Bit-rate scalability in LPC-ABS wideband coder
M1550Bernhard Grill Update of the preliminary MPEG-4 Working Draft

The following documents were also noted as relevant to the business of Audio Subgroup
NumberSection SourceTitle
M1533Requirements Rob KoenenReport of AHG on MPEG-4 requirements
M1454SNHC Ronnie R. BurnsResponse to CFP on SNHC Audio
M1472SNHC Itaru KanekoChairman's report of Ad-hoc group on the SNHC Audio
M1438Test Laura ContinReport of the ad hoc group on MPEG-4 July '97 audio visual tests
M1440Test Laura ContinTest methods and procedures for July '97 MPEG-4 tests
M1484Test C.S.Boon, M.Okuno,

S.Kadono

Audio-visual Sequences for July 97 Tests

18. Output Documents

The following output documents were produced by the Audio Subgroup. Those shown in Italics were approved for public release.
NumberAuthor Title
N1419David Kirby (BBC), Kaoru Watanabe (NHK) Report on the Formal Subjective Listening Test of MPEG-2 NBC multichannel audio coding
N1420David Kirby (BBC), Kaoru Watanabe (NHK) Overview of the Report on the Formal Subjective Listening Test of MPEG-2 NBC multichannel audio coding
N1421D. Thom MPEG-4 and MPEG-2 NBC Decoder VM Software requirements
N1424M. Coleman, D. Thom et. al Initial draft of NBC Conformance and Technical Report Work plan
N1432D. Thom Call for new audio test sequences for MPEG-4
N1433G. Stoll Audio FAQ for the MPEG Home Page
N1434G. Stoll Proposed changes to ISO/IEC DIS 13818-3 [Revision of first edition (ISO/IEC 13818-3: 1995)
N1455Convenor Liaison statement to ITU-R Working Party 10C
N1456Convenor Liaison statement to DAVIC
N1457Convenor Liaison statement to ITU-R Study Group 11
N1482G. Stoll Status of Copyright Release for Audio Conformance Test Bit Streams
N1503Grill, B. et. al. Software decoder VM for MPEG-4 audio
N1504K. Brandenburg Core experiment methodology for MPEG-4 audio

19. Recommendations

A list of recommendations was prepared for approval at the final MPEG plenary meeting.

20. Agenda for next meeting

The agenda for the MPEG Audio Subgroup meeting in February 1997 in Sevilla, Spain was approved (see xxxxx).

21. Close of meeting

Mr. Schreiner thanked the participants for all their hard work in preparation for and during this meeting. He also thanked Mr. Meares for acting as Subgroup Secretary, and the hosts for the facilities provided. With that, he declared the Audio Subgroup meeting closed and wished members a safe return journey.


Annex A-I

37th MPEG/Audio Maceio Meeting Participant List (November 1996)

NameCountry Affiliation e-mail address
Akagiri, K.J Sonyken@av.crl.sony.co.jp
Baroncini, VI Fondazlone Ugobordoni vittorio@fub.it
Bosi, M.USA Dolby Laboratories mab@dolby.com
Brandenburg, K.DE FhG - IISbdg@iis.fhg.de
Burns, R.USA Hughesrburns@hitchcock.dcf.scg.hac.com
Coleman, M.USA FiveBatsmc@fivebats.com
Dietz, M.DE FhG-IISdiz@iis.fhg.de
Edler, B.DE University Hannover edler@tnt.uni-hannover.de
Feige, F.DE Deutsche Telekom AG feige@fz.telekom.de
Fukuchi, H.J Nippon Steel Corp fukuchi@lsi.nsc.co.jp
Gill, AharonIs Zoranaharon@zoran.co.il
Grill, B.DE Univ. of Erlangen grl@lte.e-technik.uni-erlangen.de
Hamilton, J.USA General Instruments jhamilton@gic.gi.com
Herre, J.DE FhG-IIShrr@iis.fhg.de
Iwadare, M.J NECsc29a@dsp.cl.nec.co.jp
Johnston, J.USA AT&Tjj@research.att.com
Kaneko, I.J ASCIIitaru-k@ascii.co.jp
Keith, M.USA Intel Corp.mike-keith@ccm.jf.intel.com
Kerkhof, L. v. d.NL Philips CEkerkhofl@ce.philips.nl
Kim, Y.B.KR Samsungkimyb@dspsun.sait.samsung.co.kr
Koike, T.J Sonykoike@av.crl.sony.co.jp
Lindqvist, M.S Ericssonmorgan.lindqvist@era-t.ericsson.se
Lueck, C.USA TIlueck@hc.ti.com
Mainard, LFR CCETTlmainard@ccett.fr
Matsumoto, J.J Sonyjun@pcrd.sony.co.jp
Meares, D. J.UK BBCdavid.meares@rd.bbc.co.uk
Miki, S.J NTTmiki@splab.hil.ntt.co.jp
Moriya, T.J NTTmoriya@splab.hil.ntt.co.jp
Naka, N.J NTTDoCoMompeg4@mlab.nttdocomo.co.jp
Nishiguchi, M.J Sonynishi@pcrd.sony.co.jp
Nomura, T.J NECsc29a@dsp.cl.nec.co.jp
Norimatsu, T.J Matsushitanorima2@arl.drl.mei.co.jp
Oikawa, Y.J Sonyoikawa@av.crl.sony.co.jp
Ojala, P.FIN Nokia Research Centre pasi.ojala@research.nokia.com
Oomen, W.NL Philipsoomena@natlab.research.philips.com
Parladori, G.I Alcatel Telecomgparladori@tlt.alcatel.it
Puellen, RDE Pultituderpuellen@aol.com
Quackenbush, S.USA AT&Tsrq@research.att.com
Schreiner, P. G.USA Scientific Atlanta pgs@sciatl.com
Spille, J.DE Thomson Multimedia spillej@tcernd1.hanover.tce.de
Stoll, G.DE IRTstoll@irt.de
Tanaka, N.J Matsushitanatanaka@telecom.mci.mei.co.jp
Taori, R.NL Philips Researchtaori@natlab.research.philips.com
Thom, D.USA Mitsubishidthom@msm.mea.com
Väänänen, M. FINNokia Res. Center mauri.vaananen@research.nokia.com
Watanabe, K.J NHKwatanabk@strl.nhk.or.jp
Wen-Yuan, ChenTW CCL/ITRIwychen@m2sun3.ccl.itri.org.tw
Yoshida, K.J Matsushitakyoshida@telecom.mci.mei.co.jp


Annex A-II

Agenda for the 37th MPEG/Audio Subgroup Meeting

in Maceio, Alagoas, Brazil, November 1996
1) Opening of the meeting

2) Approval of agenda

3) Allocation of contributions

4) Communications from the Chairman
- Joint meetings (Test, Requirements)

5) Tampere and Chicago meeting reports

6) Report of ad hoc group activities / M1374, M1378, M1396, M1424, M1442, M1448, M1449, M1472

7) Resolution of National Body comments / M1474, M1574, M1443

8) MPEG-2 BC
- MPEG-2 audio quality update
- DIS 13818-3.2
- Bitstream tests Copyright release for bitstreams

9) MPEG-2 NBC
- Results of multichannel core experiments / M1383
- NBC formal multichannel test results / M1371, M1436
- NBC RM development / M1372
- NBC DIS preparation /M1378, M1574
- NBC technical report (software release)
- NBC compliance
- NBC syntax and MSDL
- NBC name change

10) MPEG-4
- VM activities / M 1408, M1505, M1523
- Requirements and Working Draft / M1533, M1550
- SNHC / M1454, M1464, M1477, M1389
- Other developments / Software release/ M1418, M1437, M1473, M1518
- Complexity
- Core experiments / M1435, M1508, M1509
- July 1997 tests / M1438, M1440, M1484

11) Availability of m-channel test material

12) Preparation of a press statement

13) Liaison matters
- ITU-R, DAVIC - NBC Test results

14) Discussion of unallocated Contributions

15) Recommendations for final plenary

16) Agenda for next meeting

17) A.O.B.
- FAQ on MPEG Audio

18) Closing of the meeting


Annex A-V

Audio Task Groups

NBC DIS preparation - Bosi
BosiColemanLueck OikawaQuackenbush
HerreIwadareFukuchi BrandenburgDietz
JohnstonAkagiriKerkhof

MPEG-4 WD editing - Grill
GrillBurnsNishigushi MainardMoriya
NakaMikiLindquist SpilleFeige
MatsumotoKimYoshida TanakaNorimatsu
BrandenburgHerreKoike KanekoStoll
AkagiriVaananenEdler Taori

MPEG NBC Technical Report and Compliance - Coleman
QuackenbushThomDietz JohnstonColeman
AkagiriOikawaKerkhof

MPEG Audio FAQ - Stoll
StollKerkhofJohnston QuackenbushKaneko
ParladoriThomKerkhof BrandenburgBosi
FeigeNishigushiAkagiri Oomen

MPEG-4 Requirements - Edler
AkagiriBurnsEdler Kerkhof Koike
MatsumotoMoriyaNaka NishigushiTaori
LindquistMikiStoll

Core Experiments - Brandenburg
AkagiriBrandenburgEdler Kerkhof Koike
MatsumotoMoriyaNaka NishigushiNorimatsu
OjalaOomenStoll TanakaTaori
VaananenYoshida

Editing of NBC Test results Web document - Meares
MearesSchreinerWatanabe BaronciniSuzuki
David Kirby by phone and email


MPEG-4 SNHC Final Meeting Report

Pete Doenges

15 February 1997

Report

The objectives of the SNHC working groups at Maceió were to:

The results of the meeting included:

On Sunday, the Video SNHC AHG met with other SNHC (FBA, MITG), Systems, and Video groups to understand better what SNHC requirements were for Video, and in what cases the Video VM and tools might be used or adapted to provide the needed functionality. Presentations were quite useful in describing Video functionalities to new SNHC people, and in providing SNHC application examples with which to challenge baseline Video capabilities. This was a very useful session that showed where Video expertise and tools could be leveraged, and also where Video technology could be extended, particularly in applications of SNHC where unusual manipulation and scaling of static texture or video was required. Video experts called for more careful statement of SNHC requirements, and also began to see opportunities to re-examine texture coding and scene composition issues. By Tuesday, a combination of insights from this meeting and of SNHC texture/video meeting contributions spawned a Texture working group, jointly linked to Video and SNHC, to begin work on VM/CE designs. This group later made important contributions to SNHC requirements and a texture portion of SNHC VM/CE work.

Early in the week the following contribution presentations were given:

Face/Body Animation
M1569Experiments of FAPs Coding Tao, Huang, Chen, et al UIUC/Rockwell
M1583Facial Texture Coding Li, Davoine, Forshheimer Linkoping U.

Spatial/Synthetic Audio
M1389Music Synthesis for MPEG-4 SNHC FujiiYamaha
M1437Human Sound Localization SpilleThomson
M1454NetSound: Real-Time Audio from Semantic Descriptions via Burns/HughesMIT
M1524MPEG-4 TTS Interface Hahn, LeeETRI

Sprite & Mesh Objects (Sunday)
M1545Insight on Sprite Technology and Relation to Video and SNHC Chen, Lee, GuMicrosoft
M1567Revised Syntax and Results for CE M2 (Triangular Mesh-based Coding) Beek, TekalpU. Rochester

Texture Coding & Scaling
M1475Efficient Texture Coding Using SVQ Torres, GimenoPolytechnic U. of Catalonia
M1501Viewer-dependent Texture Coding for Transmission of Virtual Environment Jordan, EbrahimiEPFL

SNHC VM
M1477Draft Specification of SNHC VM 1.1 Horne, EbrahimiMediamatics

During the week, the greatest effort went into the SNHC VM editing work (meeting document N1454) and the agreement on details of Core Experiments and partnering in the four areas. The SNHC VM was published after the meeting without development of the Texture Coding Section, due to the recent launch of that group and the need for more time to develop a Texture VM document. SNHC groups agreed that the Texture Coding VM document and software would be consolidated and joined with the SNHC VM during Sevilla. Application scenarios for all work items were developed with test sequences for the CEs. Core Experiments are:

1. Face & Body Animation (N1452)

CE1: FAP Quantization & Coding (Virtual Meeting)

Future:

CE2: Static Facial Mesh Compression

CE3: Facial Texture Compression

2. Media Integration of Text & Graphics (N1450)

M1: CEs on Media Integration of Text

M2: CEs for Media Integration of Images & Graphics

3. CEs in SNHC Audio (N1451)

HSTTS: Prosodic Speech & Facial Animation

4. CEs in SNHC Texture Coding (N1453)

X1: Efficient Coding of Homogeneous Textures (Stochastic VQ)

X2: Facial Texture Coding (Facial Eigentextures)

X3: View-Dependent Texture Coding for Transmission of Virtual Environment

X4: Mesh-Based Texture Mapping for Synthetic Transfiguration & Augmented Reality

X5: Efficient Coding of Static SNHC Texture to be Mapped in 2D/3D Surface Models

The SNHC Audio group also presented important documents, and audio demonstrations, associated with absent submitters, Yamaha and MIT. These contributions offered considerable promise of coding synthetic sound and music consistent with SNHC objectives. A careful effort was to be made to enlist the submitting organizations and others for the Sevilla meeting.

Considerable discussion among all SNHC working groups was held on the issue of intellectual property rights and about the sessions and draft document on licensing and copyright of software used in MPEG VMs. After carefully segregation of bitstream syntax and encoding/decoding technology from specific proprietary models that some members use in CEs, members finally agreed to an open software (subject to the licensing/copyright statement) for SNHC VMs.

During the week, a meeting was held between Systems/MSDL Architecture experts and SNHC. This meeting compared SNHC requirements for spatial/temporal scene composition with compositing functionality of the Video VM and with a couple different options for extending or combining MSDL with VRML. These options include using the VRML EAI as a mechanism for accessing VRML features without necessarily achieving optimum performance or functionality. Another option is to extend the MSDL class architecture to encompass a likely subset of VRML. Julien Signes gave a very useful overview of VRML to people less familiar with its design, and he compared its object abstractions with those of the MSDL class hierarchy. The approaches appeared to satisfy many SNHC requirements. However more study was needed, and contributions for Sevilla were expected that could characterize this more conclusively.

The focus during and after Chicago work on the SNHC VM was bitstream syntax, the decoded representation of the data, the architecture for decoders, and their linkages for specific hybrid applications like face animation with HSTTS. The work in Maceió, spurred by the development of Core Experiments with partners for testing, turned more to encoding/decoding algorithms and application scenarios for the CEs. Maceió marked a turn to software VM development. SNHC experimenters needed before Sevilla agreement on how software VMs would be developed and shared to make tests comparable. Individuals needed to establish rules for the software architecture for their VM developments. Discussion was held with MSDL experts about whether a practical cut-over to MSDL should be made at that point by SNHC people. Approaches were sufficiently diverse, combined with a need for flexibility and speed of development, that SNHC VM developers decided not to use MSDL yet. This must be addressed at Sevilla.

A meeting was held with Requirements to review the baseline or Extension_0 of SNHC requirements from N1316, as subsequently elaborated, with the developments in SNHC of FBA, MITG, Audio, and Texture. Major efforts were made subsequently during the week by all four working groups to assemble more meaningful SNHC requirements in a standard format. Results were turned over to the Requirements group and incorporated.

Participants and contributors to SNHC sessions and related work in Maceió
NameCompany/Lab/Univ. E-Mail
Chieteuk AhnETRI ahnc@video.etri.re.kr
Ronnie BurnsHughes Electronics rburns@hitchcock.dcf.scg.hac.com
MichaelMIT Media Lab mkc@media.mit.edu
Homer ChenRockwell Intíl homer@risc.rockwell.com
F. DavoineLinkoping Univ. davoine@isy.liu.se?
Filip DefoortKULeuven/ESAT/Visics Filip.Defoort@esat.kuleuvan.ac.be
Bert DeknuydtKULeuven/ESAT/Visics deknuydt@esat.kuleuvan.ac.be
Peter DoengesEvans & Sutherland pdoenges@es.com
Touradj EbrahimiEPFL ebrahimi@desun1.epfl.ch
R. ForshheimerLinkoping Univ. robert@isy.liu.se?
Shigeki FujiiYamaha Corp. fujii@edc3.yamaha.co.jp
Andreas Graffunder Deutsche Telekomgraffunder@fz.telekom.de
Minsoo HahnETRI mshahn@audio.etri.re.kr
Caspar HorneMediamatics caspar@mediamatics.com
Frederic JordanEPFL jordan@ltssg3.epfl.ch
Itaru KanekoASCII, Ltd. itaru-k@ascii.co.jp
Jiro KattoNEC/Princeton Univ. katto@ee.princeton.edu
Ajay LuthraGeneral Instruments aluthra@gi.com
Iole MoccagattaTexis Instruments iole@hc.ti.com
Ram NagarajanHughes Electronics ram@hitchcock.dcf.scg.hac.com
Shinya NakajimaNTT nakajima@nttspch.ntt.jp
Joern OstermannAT&T Laboratories ostermann@research.att.com
Eric PetajanLucent Bell Labs edp@bell-labs.com
Roberto PockajDIST - U. of Genoa pok@dist.unige.it
Ganesh RajanGeneral Instruments grajan@gi.com
Radu S. JasinschiTektronix radu.s.jasinschi@tek.com
Eric ScheirerMIT Media Lab eds@media.mit.edu
Takanori SenohMatsushita seno@drl.mei.co.jp
Ibrahim SezanSHARP Laboratories sezan@sharplabs.com
Paris SmaragdisMIT Media Lab paris@media.mit.edu
Jens SpilleThomson MultiMedia spillej@tcernd1.hanover.tce.de
Murat TekalpUniv. of Rochester tekalp@ee.rochester.edu
Luis TorresPolytechnic U. Catalonia luis@gps.tsc.upc.es
Barry VercoeMIT Media Lab bv@media.mit.edu

Absent on-going contributors:
Tolga CapinEPFL capin@lig.di.epfl.ch
Fabio LavagettoDIST - U. of Genoa fabio@dist.dist.unige.it
Igor PandzicUniv. of Geneva Igor.Pandzic@cui.unige.ch


Annex 8

Report of Test meeting

Source: Laura Contin, Chair

1. Introduction

The main goal of the Test Subgroup meeting held in Maceio, during the 37th MPEG meeting, was to progress the specification of test conditions for the MPEG-4 July '97 tests. Though the goals of these tests had already been discussed and specified during previous meetings and within the ad hoc groups, at the Maceio meeting it came out that further clarification was needed. Thus, document WG11/N1441 was issued.

Summarising, the goals of the MPEG-4 July '97 tests are:

Verification has the purpose of assessing the status of the work, getting feedback, and demonstrating the potential of MPEG-4.

This refers to carrying out a test with the goal to use the output of the test for a decision about whether to include the technology in the standard. A previous request of the Implementation Subgroup was withdrawn, because decisions can be taken by core experiment process. Since no further requests for these tests have been presented, these tests will not be addressed in this report and it is likely they will not be included in the July '97 tests.

By mean of this evaluation, the new proposals will be ranked according to their performance and they will be compared to the VM, whenever appropriate.

Test Subgroup worked in close collaboration with the Audio and Video Subgroups and this led to considerable progress, as is illustrated in the sections below. Test Subgroup also defined a new audio-visual library and reviewed the NBC audio test report.

2. MPEG-4 Input library

At the Chicago MPEG meeting (September '96) it was recognised that there was a need for long video sequences and new audio-visual sequences.

In Maceio, new tests sequences were proposed. Most of them were included in the MPEG library. One sequence from Microsoft was excluded because the situation about copyright was not clear. One sequence from CCETT was excluded because the segmented objects did not correspond to "real" objects.

The characteristics of the sequences included in the MPEG-4 input library are summarised in the table below:
Name of Sequence
Kind of

signal
Kind of material
Source format
Duration

(sec)
Language
AV correlation
Class
# of mask available
Akiyo

(Matsushita)
A&V
Hybrid
ITU-R601

(60 Hz)
22
Japanese
A&V correlated
A
2
Sean

(Matsushita)
A&V
Hybrid
ITU-R601

(60 Hz)
21
English
A&V correlated
A
2
News

(Matsushita)
A&V
Hybrid
ITU-R601

(60 Hz)
43
Japanese

English
A&V correlated
B
3
Weather

(Matsushita)
A&V
Hybrid
ITU-R60

(60 Hz)
32
Japanese
A&V correlated
A
2
Fish

(Matsushita)
Video
Hybrid
ITU-R601

(60 Hz)
20
N/A
N/A
E
2
Akiyo_TV

(Matsushita)
A&V
Natural
ITU-R601

(60 Hz)
25
Japanese
A&V correlated
A
N/A
Sean_TV

(Matsushita)
A&V
Natural
ITU-R601

(60 Hz)
68
English
A&V correlated
A
N/A
AT&T-1
A&V
Natural
ITU-R601

(60 Hz)
30
English
A&V not correlated
C
N/A
AT&T-2
A&V
Natural
ITU-R601

(60Hz)
30
English
A&V part. correlated
C
N/A
AT&T-3
A&V
Natural
ITU-R601

(60 Hz)
30
English
A&V part. correlated
C
N/A
Telefonica
A&V
Hybrid
ITU-R601

(50 Hz)
420
English
A&V part. correlated
A/B(*)
N/A
CCETT
A&V
Natural
ITU-R601

(50 Hz)
420
English
A&V correlated
A/B (*)
N/A
Microsoft
V
Synthetic
ITU-R601

(60 Hz)
10
-
-
B
N/A

(*) The sequence is quite long and includes parts that belong to different classes.

Test material (both video and audio-visual) to be used during the July '97 tests should be selected from these new sequences.

3. July '97 verification tests

The verification tests have the purpose of assessing the status of the work, getting feedback, and demonstrating the potential of MPEG-4. In the context of these evaluations, the Test Subgroup has been required to carry out the following tests:

  1. formal audio-visual subjective tests for evaluating the performance of MPEG-4 in real-time communications
  2. formal video quality tests for evaluating the robustness of the MPEG-4 video VM in error-prone environments.

3.1 Evaluation of the MPEG-4 real-time communication profile

The real-time communication profile is the first profile defined by MPEG-4 and the most comprehensively described. Thus, the verification tests, that will be basically tailored on specific profiles, will start with this profile. A comprehensive evaluation of this profile, that should take into account for example the impact of the global delay in a free conversation, would require a real-time codec. Since this will not be available in time, in the July '97 tests only one-way tests will be carried out.

In these tests, audio and video coding efficiency will be jointly assessed, while global delay will be evaluated only by experts. Also for these tests, a version of the VM with rate control is required and an implementation of the ITU-T H.324 will be used to produce anchor conditions.

The bitrates that will be tested are specified in the table below:
MEDIA

TYPE
TOTAL BITRATE
AUDIO BUDGET
VIDEO BUDGET
SYSTEM BUDGET
PSTN
14.4 Kbps
2 Kbps
12 Kbps
0.4 Kbps
PSTN
28.8 Kbps
6 Kbps
22 Kbps
0.8 Kbps
ISDN
64 Kbps
16 Kbps
46 Kbps
2 Kbps

The ad hoc group on MPEG-4 video low-delay evaluation (chaired by M. Zeug) will provide volunteers to produce the test material, that includes both MPEG-4 and H.324 decoded audio-visual sequences.

The source sequences will be selected at the next meeting.

3.2 Evaluation of the error robustness of the MPEG-4 video VM

During a joint meeting with the task group for video error resilience, it came out that there was the possibility to carry out verification tests based on a real-time decoder and using a software implementation of the system layer. Advantages of this configuration are:

  1. test conditions would be closer to the real situation
  2. different error patterns could be used
  3. the time for producing test conditions would be considerably reduced (decoding and editing the D1 tapes would not be needed anymore)

The ad hoc group on error resilience (chaired by J. Brailean) will study the feasibility of this solution.

3.3 Other tests

Members of the Video Subgroup asked the Test Subgroup to carry out tests on stereo sequences and tests on critical sequences at bitrates higher than 1 Mbit/s. These tests will be discussed further. If they are not be feasible by July '97, they will be included in the plan of the verification tests that should start from March '98.

The Audio Subgroup has not yet asked for any specific verification tests, though a few proposals are under discussion (e.g. multichannel and object-oriented tests).

4. July '97 competition tests

By mean of this evaluation the new proposals will be ranked according their performance and they will be compared to the VM, whenever appropriate.

In order to manage the logistical questions related to these tests, the Companies that intend to submit new proposals either for audio, video or audio-visual tests must send pre-registration forms to the chairmen of the Test and Requirements Subgroups. The conditions for the submission are illustrated in the Call for proposals (WG11/N1499) while the pre-registration forms are attached to the evaluation document (WG11/N1442).

4.1 Video tests

It was previously agreed that, in the Competition Tests, the functionalities to be tested are the same as those tested in previous MPEG-4 tests (i.e. coding efficiency, error robustness, object based manipulation that also includes spatial and temporal scalability). It was also proposed to use different test methods for error robustness and object based manipulation (doc. WG11/1310, July '96). At the Maceio meeting the discussion was mainly focused on the test conditions.

During a joint meeting with the Video Subgroup general issues were discussed. The problem of the filters for up- and down-conversions was raised. It was agreed that the up- and down-sampling filters defined by the VM will be used to produce test conditions coded by means of the VM at all the bitrates indicated in the evaluation document, unless better filters will be proposed for the highest bitrates. The proposers will be required to use the same filters.

Additional issues discussed were: anchor conditions, frame rates and bitrates. Concerning the first point, no agreement was reached. Concerning the test conditions, there was a suggestion to define a minimum frame rate, but it has still to be specified.

During a joint meeting with the Video error resilience task group, it was decided to use the error patterns specified in the evaluation document for evaluating the performance of the new technologies against the VM.

No particular progress has been achieved for object based manipulation tests. The Test Subgroup requires the Video Subgroup to specify the values of the following parameters:

4.2 Audio tests

During a joint meeting with the Audio Subgroup, it was agreed that, for testing new audio technologies against the VM, basically the same methodology used in the last MPEG-4 audio tests will be adopted.

A few modifications were discussed and they can be summarised as follows:

  1. New audio excerpts are needed, because using the available material could bias the test results. A call for new audio test material was issued (N. 1432). The characteristics of the material required are:
    1. English or German speech, also with background noise or music
    2. Music items of different levels of complexity (one instrument, multiple instruments)
    3. Pre-segmented items for demonstration of object-oriented functions
    4. Stereo
  2. Concerning the definition of the test conditions, the following points were agreed:
    1. G.723 with 6.3 kbit/s is a more appropriate reference codec for 6 kbit/s.
    2. The definition of an MNRU which is less item-dependent for 48 kHz sampling might be useful. In case it cannot be made available, the MNRUs used in the previous tests will be used.
  3. The following points about specific functionalities require further discussion:
    1. In error resilience tests, MNRU at less than 17 dB should be included, error conditions should be adjusted to the requirements, more than two error patterns for each sequence should be used to reduce the effect of bit error position dependencies (in particular for lower BER conditions) and packet loss should be included, according to the proposal given in doc. M1179.
    2. Multichannel and object oriented tests could be added. At least in the case of the multichannel test, loudspeakers should be used.
    3. Test conditions for scalability should also be adjusted to the requirements, if there are any specific ones.
  4. Tests with Japanese speech and English/German speech items should be carried out at different laboratories, but preferably only one laboratory for each type of speech item. If necessary, a further separation could again be based on the reference sampling frequencies.
  5. No expert listeners should participate in the tests. If test items are to be repeated, at least 20 subjects should participate to the test. If repetition are not to be used, twice the number of subjects would be needed.
  6. In the presentation of the results, reference codecs and MNRUs should be included in all the graphics of each functionality at the corresponding bit rates, i.e. they must be included in more than one ranking group. In the ranking itself MNRUs might be omitted.

5. July '97 workplan

The following table summarises the workplan for the July '97 tests:
January '97
  • Definition of test conditions for error robustness verification test (choice between using models for residual errors or use implementation layer and real-time decoder)
  • Definition of the anchor conditions for error robustness, if any
  • Definition of test conditions for AV tests for the real-time communication profile
10 February '97
  • Pre-registration
February MPEG meeting
  • Specification of A, V and A-V material
  • New version of the evaluation document (rev. 3)
  • Identification of potential test sites
April

MPEG meeting

  • Final version of the evaluation document (rev. 4)
  • Distribution of the work for preparing test material and setting up the tests
  • Identification of test sites
15 May '97
  • Registration
2 June '97
  • submission deadline
Beginning of July
  • Test completed
July

MPEG meeting

  • Test results

6. NBC test

During a joint meeting with the Audio Subgroup, the results of the formal listening tests of the MPEG-2 NBC multichannel codecs were presented. A few changes to the first version of the test report (doc. M1371) were agreed and included in the final version (doc. WG11/ N1419). An overview of the preparations for, and the results of, these tests was prepared to be included in the MPEG Web pages.

The NBC tests represent a big achievement for the Audio Subgroup , both for the organisation of the tests and for the results, that confirm a high performance of the multichannel NBC coding schemes.


Annex 9

Report of Implementation Studies meeting

Source: Paul Fellows, Chairman of ISG

1. Overview

There were fewer than normal attendees present. During the meeting, the group concentrated its efforts around the following topics :-

2. Video VM padding complexity review.

It was the strong opinion of the group that the move towards a macro-block based scheme will significantly reduce the amount of processing and memory bandwidth requirements. Discussions took place with members of the Video Group which resulted in three schemes being proposed to solve the problem that was identified at the Chicago meeting by Implementation Studies. The group recommended that an ad-hoc group be established to compare the schemes.

3. Concurrency Model

The group reviewed a number of contributions that discussed how best to implement an MPEG-4 terminal such that multiple objects could be simultaneously processed in a fair and isochronous manner. The contribution documents discussed these issues in precise detail. The group after reviewing the contributions recommended that the system architecture should not preclude an implementation based upon either multi-tasking or multi-processing technologies. It was felt that as the system was handling multiple simultaneous objects, each with its own distinct properties, the best implementation strategy would be to model the same behavior i.e. use of concurrency. Implementers however have the choice to implement the system in a purely sequential manner if this meets the needs of their application. Specific problems that may arise however, would pertain to such an implementations coexistence with other applications running on the same machine.

4. Video Test Materials

The group was concerned that the test materials being used by MPEG-4 were possibly unrepresentative of the functionality under test and also the computation profile expected. Specifically the use of QCIF resolution sequences were not felt to be particularly challenging for computation profile analysis. The group recommended that sequences be produced that :-

5. Computational Graceful Degradation

The work of the ad-hoc was reviewed during the meeting. This included viewing of simulations of the approach that were recorded onto a D1 tape. The techniques being developed look promising and provide for either :-

The group recommended that this activity be continued.


Annex 10

Report of DSM-CC Meeting

Source: Vahe Balabanian, DMIF Chairman

1. Assignment of a New Chairman

Vahe Balabanian

balabani@nortel.ca

Phone: +1 (613) 763-4721

Fax: +1 (613) 763-8904

Nortel

replaces the outgoing Chairman Chris Adams of Divicom. Many thanks to Chris for his untiring efforts in giving birth to DSM-CC which we are all proud of.

2. Contributions list
Doc #TitleContact Relevance
M1381Report on DSM-CC-2 Ad Hoc Vahe Balabanian
M1397Implications of full integration of DSM-CC with ATM in DSM-CC-2 Vahe Balabanian
M1398ITU-T IN Interworking with DSM-CC-2 Vahe Balabanian
M1399Multimedia Conferencing in DSM-CC-2 Vahe Balabanian
M1400Current Status of ITU-T T.130 "Real Time Audio-Visual Control for Multimedia Conferencing - Architecture & Overview Vahe Balabanian
M1401Liaison on the definition of DMS-CC PassThru type values for T.120 Vahe BalabanianDSM-CC IS
M1420DSM-CC Conformance Manfred Schumann
M1543Digital Studio Command and Control Proposed Activity Christopher Ward
M1592Report on the ad hoc final editing of DSM-CC IS Matt GoldmanDSM-CC IS
M1586Liaison to MPEG-4 re Multiplex ITU-T SG15 LBC
M1587Liaison to MPEG on MPEG-4 Systems Issues ITU-T SG15 LBC
Mxxx1Session Protocol Feedback Jim Van LooDSM-CC IS
Mxxx2DSM-CC U-U Bug Fix resume(now) Guido FranceschiniDSM-CC IS
Mxxx3U-N Download SDL Mikael SchmidtDSM-CC IS
Mxxx4Pass-Thru SDLMikael Schmidt DSM-CC IS

3. Launching the next phase of DSM-CC, DMIF

1- DMIF encompasses the following directions

a) The concept of a common end-to-end "session" interface across multiple network provider implementations

b) Integration with multiple network technologies, such as ATM and the Internet

c) Integration with specific object domains, such as CORBA and JAVA

d) Multiple devices participating as peers within the same session

e) Fully symmetric consumer and producer operations within a single device

f) The scheduling and real-time switching/multiplexing of bit streams

2- Prepared a SC29 work statement to launch DMIF N1429

a) Preliminary Working Draft -- July 1997

b) Working Draft -- February 1998

c) Committee Draft -- July 1998

d) Draft International Standard -- February 1999

e) International Standard -- July 1999

3- Prepared a press release on DMIF

4. DSM-CC, MPEG-4 Systems and MPEG-4 requirements Joint Meeting

a- N1399 Requirements for Signaling was reviewed by Carsten Herpel and potential DSM-CC contributions discussed

b- Establish a joint MPEG-4-DSM-CC AHG chaired by Carsten Herpel herpelc@tcernd1.hanover.tce.de from Systems and co-chaired by Jim Van Loo jvanloo@frstprsn.Eng.Sun.com of DSM-CC

c- The mandate is to sort out MPEG-4 requirements to Systems and DSM-CC requirements

d- Make use of application profiles (top down view) defined by the MPEG-4 Requirements

e- A subscription list has been collected contact Carsten Herpel

5. Complete 13818-6 IS Editing

1- Reorganized the Introduction section in Chapter 1 to better highlight the key features of DSM-CC

2- Fixed the U-U bug resume(now) no corrigendum is required

3- DSM-CC IS is being released to ISO for publication

4- Thanks to Matt Goldman for his supreme effort in coming through with this document

6. DSM-CC Conformance

1- M1420 was reviewed and found to be a good base for a conformance document

2- Need to add operation Conformance to M1420 in order to get it into a Preliminary Working Draft stage

3- Tim Addington Tim.Addington@Sciatl.com is tasked with coming up with a PWD in 2/97 Seville

4- It is desirable to share the output with DAVIC in Hong Kong for common text on DSM-CC parts

5- 2/97 PWD, 4/97 WD, 7/97 CD, 2/98 DIS, 7/98 IS

6- Send Request for DSM-CC Conformance Test participation 2/97

a) Bellcore

b) NIST

c) Columbia University

d) Jupiter,

7- Identify any forum or organization who might review the conformance document and participate in the process.

7. Establish a Process to market DSM-CC

7.1 Establish AHG group to write white paper and FAQ on DSM-CC

- Vahe Balabanian will put Nortel IEEE article on the dsmcc.divi.com FTP site as a preliminary white paper

- Participation will be solicited by Vahe on the dsmcc.divi.com reflector

- Input to FAQ will be requested in the feedback to the Preliminary White Paper

- Add White Paper and FAQ to MPEG sites (Tim Addington)

7.2 Update DSM-CC seminar slides

- Matt Goldman will send the existing seminar slides to Vahe, Jim and Tim for update and feedback

7.3 DSM-CC IS needs increased exposure to other standards and Forums besides DAVIC, DVB and ITU-T SG8

1- Give DSM-CC Seminars

2- Harmonize DSM-CC interworking with other standards

a) Liaison with ATSC (Advanced Television Systems Committee) a liaison is drafted and will be carried to ATSC by Tim Addington

b) Liaison with IETF Session Protocol and Real Time Stream Protocol, Jim will get in touch with expert in IETF for IETF application notes contribution to DSM-CC in 2/97 on how to bridge to the IETF complementary protocols

c) Liaison with ITU-T SG16 H.323, H.324, H.245, harmonization contributions are required to DSM-CC in 2/97

7.4 Submit overview papers to workshops

8. Liaisons

8.1 Liaison reply to SG8 (N1465)

codes (0x0001 - 0x0015) for T.120 use in Table 12-12 User-to-Network Pass-Thru type values in the DSM-CC IS Section 12.7 Pass-Thru Codes.

8.2 Liaison statement to SG8 and SG15 (N1464)

Invite SG16 to a DSM-CC Seminar at the next MPEG 2/97 meeting

8.3 Liaison to ATSC T3/S13 (N1463)

Sharing information on DSM-CC IS broadcast download


Annex 11

Report of the Liaison meeting

Source: Barry Haskell, Chair

The Liaison group considered input documents

MPEG96/1585 from ITU-T LBC group regarding requirements for the real time conversation profile. Input was considered by Requirements.

MPEG96/1586 from ITU-T LBC group regarding low delay requirements for video and systems. Input was considered by video.

MPEG96/1587 from ITU-T LBC group regarding Systems and the requirements for interoperability with H.323 and H.324. APIs were added to the VM and WD.

MPEG96/1588 from ITU-T LBC group regarding System Multiplexing and the desirability of maintaining consistent with existing ITU-T network standards. The MSDL-M group considered these suggestions and will introduce a two level Multiplex.

MPEG96/1589 from ITU-T LBC group regarding recent changes in the H.263+ draft video coding standard. The video group considered this input and

MPEG96/1401 from ITU-T SG8 on the subject of carrying T.120 in the DSM-CC. The DSM-CC considered this input and assigned codepoints.

N1783 and N1784 from ISO TC46/SC9 on the subject of IPR identification. Input was considered by Requirements and incorporated into their IPR work.

MPEG96/Mxxx from SC29/WG1 on the subjects of IPR protection and cooperation with JPEG 200. Input was considered by Video and Requirements.

Five inputs from other ISO groups upon which no action was taken:

N1786 from TC100/AWG on color measurement and control.

N1789 from TC100/AWG on human interfaces.

N1790 from TC100/AWG on formats for multimedia standards.

N1791 from JTC1 SWG-GII on the Outreach effort.

N1782 from JTC1/SC22 on plans in SC22 for Java standardization.

Nine inputs from ITU-R on which no action was taken:

11/9-E on revision of BT.656-3 (old CCIR 601)

11/12-E on new recommendation for ancillary data in CCIR 601

11/15-E on CRCs.

11/27-E on post production guidelines.

11/33-E on wide screen signaling.

11/19-E on subjective assessment (old CCIR 500)

11/7-E on data access using MPEG2 PSI.

11/8-E on multiplex using MPEG2 Systems.

11/6-E on video coding using MPEG2 video.

The following output liaison documents were produced:

WG11/N1455 Liaison statement to ITU-R Working Party 10C

was produced for sending to ITU-R WP 10C on the subject of NBC audio test results.

WG11/N1456 Liaison statement to DAVIC

was produced for sending to DAVIC on the subject of NBC audio test results.

WG11/N1457 Liaison statement to ITU-R Working Party 11

was produced for sending to ITU-R WP 11 on the subject of NBC audio test results.

WG11/N1458 Liaison response from MPEG-4 to JPEG-2000

was produced for sending to SC29/WG1 JPEG 2000 on the subject of interoperability between MPEG4 Intra coding and JPEG 2000.

WG11/N1459 Liaison response from MPEG-4 to the LBC video group was produced for sending to the ITU-T SG15 LBC video group on the subject of interoperability between MPEG4 video and the evolving H.263+.

WG11/N1460 Liaison response from MPEG-2 to the ITU-T SG15 AVC video group

was produced for sending to the ITU-T SG15 AVC video group on the subject of their requested MPEG-2 part 2 Amendment 4 for an ITU video extension code.

WG11/N1461 Liaison Response to ITU-T LBC Group

was produced for sending to ITU-T SG15 LBC group on the subjects of the real time profile and low delay operation.

WG11/N1462 Liaison Response from MPEG4 Systems to ITU-T LBC was produced for sending to ITU-T SG15 LBC group on the subject of multiplexing and interoperability with ITU-T H.3XX terminals.

WG11/N1463 Liaison letter regarding ATSC Data Broadcast was produced for sending to the U.S. Advanced Television Systems Committee on the subjects of DSM and data download.

WG11/N1464 Liaison statement concerning the DMIF Initiative

was produced for sending to ITU-T SG15 and SG8 on the subject of the DMIF initiative.

WG11/N1465 Liaison statement concerning the definition of T.120 Pass-Thru type codes was produced for sending to ITU-T SG8 on the subjects of t.120 and pass through capability in MPEG2 DSM-CC.