INTERNATIONAL ORGANIZATION FOR STANDARDIZATION
`ORGANISATION INTERNATIONALE DE NORMALISATION
`ISO/IEC JTCI/SC2/WG8
`CODED REPRESENTATION OF PICTURE AND AUDIO INFORMATION
`
`Source
`
`Title
`
`Status
`
`Leonardo Chiariglione
`Chairman of Moving Picture Experts Group MPEG
`Eighth MPEG Meeting Report
`Draft
`
`ISO-IEC/JTC1/Sc2/WG8
`MPEG 89/218
`
`October1989
`
`Eighth MPEG Meeting Report
`The 8th meeting of the Moving Picture Coding Experts Group MPEG was held at
`the
`Kurihama Laboratory of Victor Company of Japan 16 to2l October 1989 and at Matsushita Electric
`Industrial Company Osaka Japan on 23 October 1989
`crucial one since the MPEG workplan called for the Video Group to preselect
`The meetingwas
`and cluster
`the video coding proposals which had been submitted in the meantime the Audio
`Group to define development groups that would enable the optiniisation of different algorithms
`without resorting to
`preselection and the Systems Ad-hoc Group to start clarifying
`the technical
`issues behind the interaction
`between moving picture images and sound Failure to keep the time
`table would have seriously prejudiced
`the chances of keeping the deadline of 1st September 1990
`for the production of the DP
`The Subjective Test Ad-hoc Group proved to be of invaluable help in progressing
`the work of
`the Video Group thanks to the perfect arrangement and professional set up of equipment managed
`by Mr Hidaka and the VLSI Ad-hoc Group had more material
`than time to analyse the video
`coding VLSI implementation studies Lastly an investigation into the likely DSM target of the next
`generation of moving picture coding standard was started by the Future DSM Ad-hoc Group
`Annex
`gives the distribution of Group meetings during the period
`The ambitious goals of the meeting were consistently reached This was due to the dedication
`and skifi of the Group and Ad-hoc Group Chairmen ProL H.-G Musmann Dr
`Le Gall Mr
`Hidaka Mr
`Kogure Mr
`Simon and Mr
`Smith and by the support given by the over
`100 participants in the meeting attendance
`list in Annex II
`The agenda was approved
`and documents were assigned to the proper items in the agenda
`Annex ifi
`In item the Chairman communicated the main facts that had taken place since the Stockholm
`meeting
`The JPEG Chairman Mr
`Wallace had supplied
`JPEG
`description of the baseline
`system MPEG89/168 which had been used by MPEG/Video proposers to include JPEG
`into their proposals
`compatibility
`Mr
`Johnson had kindly undertaken
`the task of running another ballot
`compatibility The result appears in MPEG89/169
`The VLSI questionnaire had been finalised and
`report by the MPEG Chairman
`providing some extra detail on the procedure was presented MPEG89/172
`The Director of the Radio Receiving Institute in Leningrad
`had expressed interest in the
`MPEG/Audio work MPEG89/176
`Dr Yasuda and the Chairman had attended
`meeting organised by AFNOR the French
`Interactive Applications AVI MPEG89/179
`is WG8
`National Body on Audio-visual
`document
`issued by the Convenor containing the report of that meeting and the recom
`mendation that MHAG considers
`the French proposal on AVIs
`
`for JPEG
`
`1
`
`HP 1029
`
`

`

`paper on MPEG activities was presented at 1TUCOM89
`very large conference
`on
`electronic media sponsored by the International Telecommunication Union Participants
`were particularly interested by the prospect of
`high-efficiency moving picture coding
`standard available in one year time
`The EBU had declared their willingness
`to arrange the subjective testing of the audio
`system being considered by the Audio Group MPEG89/196
`Ms
`Seguin Chairman of COST211 bis had contributed
`revised version of their tape
`format MPEG89/171 which had already been well used by JPEG and MPEG for
`possible standardisation by MPEG The Chairman suggested that MPEG might consider
`the inclusion of sound information into that format
`NWI proposal for Audio Content Architecture MPEG89/203 The
`SC18 had issued
`Audio Group was asked to consider that document and propose possible actions
`The Stockholm meeting report and document MPEG89/128 were approved
`The Audio and Video Groups and the Test VLSI Systems Ad-hoc oups reported on the
`activities carried out during the week The DSM Ad-hoc Group did not report because
`last session
`Simon of Intel and Mr
`was scheduled to take place in the afternoon of 23 October Mr
`Kogure
`of Matsushita were chosen to lead the Systems and Future DSM Groups respectively Annexes IV
`to IX give the reports as detailed below
`
`ANNEX
`
`ANNEX
`ANNEX
`ANNEX VI
`ANNEX VII
`ANNEX VIII
`ANNEX IX
`
`Source
`
`Title
`
`Le Gall
`H.-G Musmann
`Simon
`Hidaka
`
`Smith
`
`Kogure
`
`Report on MPEG/Video Meeting
`Report on MPEG/Audio Meeting
`Report on Systems Ad-hoc Group Meeting
`Report on Test Procedure Ad-hoc Group Meeting
`Report on VLSI Ad-hoc Group Meeting
`Report on Future DSM Ad-hoc Group Meeting
`
`Following the Audio Group recommendation the Chairman will write
`liaison statement
`SC18 informing them of the Audio Coding activities
`and giving at the same time information on
`parallel activities on Video which may be needed later on by SC18
`Coordination withother WG8 subgroup had already taken place inthe past e.g in the generation
`and informally with the MHAG Chairman after the
`of JPEG baseline
`system description
`Stockholm meeting After that Mr
`Szabo had had correspondence with Mr Kretz which the
`formalised into an MHAG document MPEG89/177 Though under pressure of time the
`latter
`short meeting with MHAG trace
`Systems Group had the opportunity during the week to have
`of it appears in Annex VI
`The Kurihama/Osaka meeting led to the formal establishment
`
`of Development Groups
`
`to
`
`MPEG/Audio
`
`blocks
`
`Transform Coding with overlapping
`ATT DTB FhG development group
`CNET individual
`Transform Coding with non-overlapping blocks
`Fujitsu JVC NEC Sony
`development group
`Subband Coding with less than or equal to
`subbands
`BT NTF
`development group
`Subband Coding with more than
`subbands
`CCETF IRT Philips development group
`Matsushita
`individual
`
`candidates CNET and Matsushita will decide on joining one of
`The remaining two individual
`1989
`development groups by the deadline of Friday December
`the existing
`
`2
`
`

`

`MPEG/Video
`
`Group
`
`interpolation
`
`studying mainly uni-directional prediction in combination with motion compensated
`of every second frame Mr
`Hepper Deutsche Thomson-Brandt will
`lead the
`Group and Mr
`Goizales IBM will act as regional coordinators
`Kogure Matsushita and
`for Japan and North-America
`Participants AlT BeIlcore CECAS IBM Intel GCF Fujitsu Matsushita NTF Philips
`Siemens Sony Telenorma University of Hannover
`
`studying
`
`Group
`past and future reference Dr
`prediction based on
`Szabo
`DEC will lead the Group and Mr
`Hidaka JVC will act as regional coordinator for Japan
`Participants C-cube DEC NEC JVC Mitsubishi
`
`non-causal
`
`studying the definition of common coding format and common pre- and post-process
`ing to be used by all MPEG proposals Dr Haskell ATl will
`lead the group
`Le Gall Morris
`Takeguchi
`Ligtenberg
`
`Monison
`
`St
`
`Group
`
`Members
`Giron
`
`Hidaka
`
`The next MPEG meetings will be
`11-12 December 1989 in Stockholm do Swedish Radio for Audio matters
`23-25 Januaiy 1990 place to be determined for Video matters
`15-21 March 1990 in Tampa Florida
`
`The Kuri.hama/Osaica meeting recommendations were presented by the Chairman at the WG8
`plenary MPEG891223
`
`The meetingwas closed with expressions of thanks for the wonderful support given by the hosting
`which had contributed
`Organisations
`substantially to the successful outcome of the meeting for
`the masterly work of the Group Chairmen and the dedication of all participants
`
`3
`
`

`

`Mnex
`
`Allocation of meetings at Kurihama/Osaka
`
`Plenary Video
`
`Test
`
`Audio
`
`Systems Evaluat
`
`Core
`
`VLSI
`
`DSM
`
`16/0900
`1000
`
`16/1000
`1230
`
`16/1400
`1600
`
`16/1600
`1730
`
`17/0900
`1230
`
`18/0900
`1000
`
`18/1000
`1230
`
`18/1400
`1630
`
`19/0903
`1230
`
`19/1403
`
`17.30
`20900
`1230
`
`20/1400
`1600
`
`20/1600
`1700
`
`20/1700
`1730
`
`21/0900
`1230
`
`2111400
`1600
`
`23/0900
`1230
`
`23/1400
`1730
`
`4
`
`

`

`Annex II
`
`Kurihama/Osaka Meeting Attendance Ust
`
`Sony Corp
`Mitsubishi Electric Corp
`Intel Corp
`1TF Intermetall
`
`Star Signal
`CSELT
`Bell Northern Research
`ccirr
`Victor Company of Japan Ltd
`Universitaet Hannover
`Gcr
`ASCII Corp
`Fraunhofer Gesellschaft
`Zoran Corp
`International Business Machines Co
`Matsushita Electric Industrial Co Ltd
`CNFF
`Siemens AG
`American Telephone
`and Telegraph
`Nippon Telegraph and Telephone Corp
`Deutsche Thomson-Brandt GmbH
`Victor Company of Japan Ltd
`011 Electric Co
`Matsushita Electric Co Ltd
`Ltd
`Fujitsu Laboratories
`Ricoh Company Ltd
`Ricoh Company Ltd
`American Telephone
`and Telegraph
`American Telephone
`and Telegraph
`Bell Communications Research
`Nippon Telegraph and Telephone Corp
`Toshiba Corp
`GCT
`British Telecommunications
`Matsushita Electric Industrial Co Ltd
`Victor Company of Japan Ltd
`Telefonbau mid Normaizeit GmbH
`Bell Communications Research
`CNEF
`C-Cube Microsystemsorp
`Nederlandse Philips Bedrijven
`Toshiba Corp
`CNET
`Fujitsu Limited
`CASIO Computer Co Ltd
`Yamaha Corp
`German Aerospace Research Establish
`Victor Companyof Japan Ltd
`Philips Research
`Laboratories
`British Telecommunications
`Mitsubishi Electric Corp
`Universitaet Hannover
`
`plc
`
`plc
`
`USA
`
`USA
`
`CDN
`
`USA
`USA
`
`USA
`
`USA
`USA
`USA
`
`GB
`
`USA
`
`USA
`NL
`
`GB
`GB
`
`81462306284
`81467471251
`16099367700
`49761517880
`
`39 112169520
`16137632697
`3399024098
`81468368540
`49 511 762 5333
`8134987543
`813 499 2146
`499131606330
`14089861240
`19149452141
`816 906 1748
`33 145295435
`498963648322
`12019493697
`81422593393
`49511418 2483
`81468368540
`8137987631
`81449342862
`81t4 7542580
`8145593 3489
`81455933489
`12015827308
`12015827308
`12018297019
`81469592829
`81457563529
`813 498 7543
`44473642157
`816906 1748
`81468368540
`4969 272 3315
`12018297019
`3376903443
`14154962205
`3140734995
`81457563529
`3396053530
`81447542582
`81425632456
`
`49 2203 6012866
`81468360421
`44293776495
`44473 643791
`81467471251
`49 511 7625333
`
`KAkagfri
`KAsai
`
`Astle
`
`Butera
`J.Caras
`
`Chiariglione
`Costa
`
`Deheiy
`
`Fuchigami
`Fuchs
`
`Fujiwara
`Fukutome
`
`Gerhaeuser
`Gil
`
`Gonzales
`Goto
`
`Guichard
`Hammer
`Haskell
`
`Hayashi
`Hepper
`Hidaka
`Hontama
`Inoue
`
`Iseda
`
`Ishikawa
`
`Ishikawa
`
`Jayant
`Johnston
`
`Judice
`
`Kamilcura
`
`Kaneshige
`Katayama
`Kim
`
`Kogure
`Konishi
`Kummerow
`D.LeGaII
`Le Scan
`
`Ligtenberg
`Lokhoff
`lvtMaeda
`Mahieux
`Matsuda
`
`Matsui
`
`Matsuyama
`
`Meier-Engelen
`Mon
`Morris
`
`Morrison
`Murakami
`Musmann
`
`5
`
`

`

`Nagai
`
`Nagata
`Nishitani
`
`Noll
`
`Normile
`
`Oestberg
`Ohta
`Okamoto
`Onoe
`
`Porter
`
`A.Puri
`
`Saint-Girons
`
`Schylander
`
`Sebestyen
`Simon
`
`Smith
`
`Smith
`
`Spille
`
`Stoll
`
`Sugiyama
`
`Sugiyama
`
`Sugiyama
`Suzuki
`Szabo
`
`Takahashi
`
`Takahashi
`
`Takeguchi
`Tanaka
`
`Texier
`Tsuda
`Ueno
`
`Unagami
`Uwabu
`Van der Meer
`Watanabe
`Wise
`
`Wright
`Yamada
`KYang
`Yasuda
`
`Yasura
`Yeh
`Yonemitsu
`Yung
`
`Matsushita Electric Industrial Co
`Matsushita Electric Industrial Co Ltd
`NEC Corp
`Technische Universitaet Berlin
`
`plc
`
`Institut
`
`Apple Computers
`SAFAD
`NEC Corp
`Hitachi Ltd
`Victor Company of Japan Ltd
`Cypress Semiconductor
`American Telephone
`and Telegraph
`Thomson Consumer Electronics
`Philips International B.V
`Siemens AG
`Intel Corp
`British Telecommunications
`INMOS Ltd
`Deutsche Thomson-Brandt GmbH
`fuer Rundfunktechnik
`NEC Corp
`Toshiba Corp
`Victor Company of Japan Ltd
`Sony Corp
`Digital Equipment Corp
`Matsushita Electric Industrial Co Ltd
`Victor Company of Japan Ltd
`Matsushita Electric Industrial Co Ltd
`Matsushita Research
`CNFF
`Ltd
`Fujitsu Laboratories
`Victor Company of Japan Ltd
`Ltd
`Fujitsu Laboratories
`3CF
`Philips Consumer Electronics
`Nippon Telegraph and Telephone Corp
`Inmos Ltd
`
`Institute
`
`Apple Computers
`Victor Company of Japan Ltd
`Bell Communications Research
`Nippon Telegraph and Telephone Corp
`Victor Company of Japan Ltd
`Apple Computers
`
`SonyCorp
`Apple Computers
`
`816 905 4062
`8169061748
`81448562229
`4930 314 22514
`408 974 1794a
`468 137025
`81448562239
`81 45 8644305
`81468368844
`14089432949
`1201 9493697
`1213 568 9002
`3140757646
`498972247713
`609 936 7700
`44473643791
`44 454 617910
`495114182832
`4989 32399351
`81448562229
`81445492261
`8129735605
`813 447 5989
`1508 486 7417
`81 6906 1748
`81468368540
`816 906 1748
`8144 933 1797
`33 452 5435
`8144 754 2582
`
`8144 754 2580
`8134987543
`3140758503
`81468592829
`44454616616
`14089746615
`81 468 36 8540
`12018297019
`81468592806
`81468368540
`1408 974 6615
`81462306284
`14089746615
`
`USA
`
`USA
`USA
`USA
`NL
`
`USA
`GB
`GB
`
`USA
`
`NL
`
`GB
`USA
`
`USA
`
`USA
`
`USA
`
`6
`
`

`

`Annex III
`
`Kurihama/Osaka Meeting Agenda
`
`MPEG P1enaiy
`Opening of the meeting
`Approval of agenda
`Allocation of contributions
`Communications from the Chairman
`Stockholm meeting report
`Proposal package description
`Reports and planning of subgroup activity
`Video
`Audio
`
`System aspects
`
`tests
`
`Subjective
`VLSI aspects
`Digital storage media
`Coordination with other WG8 subgroups
`Set up of Development Groups
`10 Time and place of next meeting
`11 Recommendations
`of Kurihama meeting
`12 AO.B
`13 Close of the meeting
`
`170174
`
`168 169171172173
`161
`
`128
`
`176 177179 196
`
`181-196211
`
`176197-210
`180212-213
`141215
`172
`
`214
`
`177-179
`
`MPEG/Audio Group
`Discussion of the submitted detailed descriptions of algorithnis
`Preselection
`Clustering of the proposed audio coding algorithms
`Discussion of the future procedure
`If preselection test
`the test procedure has to be finalised
`is necessaiy
`Refinement of the proposal package description
`Performance to be tested continued
`from last meeting
`Definition of test procedures
`test material
`
`test
`
`location
`
`evaluation
`
`of tests
`
`A.OB
`
`MPEG/Video Group
`Monday Oct 16th AM PM
`
`Friday Oct 20th pm
`Saturday Oct 21st am pm
`Monday Oct 23rd am
`
`Approval Stockholm Meeting Report
`Discussion Document
`128/Rev3
`Result JPEG Compatibility Ballot
`of proposals
`Presentation
`10 minute/proposa1
`Establishment verification teams
`Bit rate Random Access number
`Review Results Subjective Test
`Review Results Verification Teams
`Establishment Development groups
`Discussion of Frame rate Conversion
`9a...242530..Hz operation
`9b Inter-standard conversion
`
`7
`
`

`

`Test Group
`Tu 17 pm Explanation of assessment method
`Demonstration
`QA
`of Assessors 25 assessors will be required
`Training of assessors
`Assessment session
`We 18 am Assessment session
`Assessment session
`
`Subjective
`
`Selection
`
`We 18 pm Assessment session
`
`Assessment sessionS
`
`Assessment session
`Th 19 am Assessment session
`Assessment session
`
`Assessment session
`Th 19 pm Assessment session 10
`Assessment session 11
`
`Assessment session 12
`Fr 20 am Assessment session 13
`Assessment session 14
`
`Assessment session 15
`Fr 20 pm Assessment session 16
`
`625/50
`625/50
`625/50
`625/50
`525/60
`525/60
`525/60
`525/60
`525/60
`525/60
`525/60
`525/60
`525/60
`525/60
`525/60
`525/60
`
`MPEG/Systems Group
`Requirement spec of overall multiplex
`Types of information
`Numbers of such types
`tyming constraints between
`Basis of multiplex
`synchronous or asynchronous
`Means to locate specified parts of encoded material
`Allowed video formats
`one extreme is CIF only
`other extreme is any spetial and temporal sampling densities
`between
`are discrete solutions
`if JPEG data is on the medium then we may face the 525/625 issue not addressed in JPEG
`File structures Questions such as does the medium have one file running from start to end
`medium or are there multiple files with headers somewhere
`with MHAG
`
`these informations
`
`ie numbering indexing
`
`of
`
`Relationship
`
`VLSI Group
`Opening meeting
`Agenda approval
`Approval of Stockholm meeting report
`List of input documents Questionnaire result and VLSI proposals
`Generation of results of VLSI evaluation
`
`DSM Group
`Present status and future aspects of picture coding
`Communication network
`Picture and sound coding
`Optical data storage media
`Tape data storage media
`Other data memory
`Application of picture coding
`Layer consideration
`Picture quality consideration
`
`8
`
`

`

`Annex IV
`
`Separate Meeting of the MPEG/Video Group
`
`Source
`
`Status
`
`Didier Le Gall
`chairman MPEG-Video
`Draft
`
`MPEG-Video met inJapan first in Kurihama from October
`16 to October2lst and then inOaka
`on October 23rd This series of meeting of MPEG-Video was
`very important milestone in the
`path to achieving
`standard for video compression on digital storage media During the Kurihama
`meeting 15 proposals
`for video coding were thorougly analysed verified and tested including
`subjective testing and the commonality and differences between the different algorithms proposed
`was established
`During the Osaka meeting method was devised to evolve the results of the Japan
`meeting first into two reference algorithms then after further discussion and testing toward
`Standard ProposaL
`
`draft
`
`1/ Kurihama meeting Venfication and Testing
`
`describing the algorithm Proposal were submitted to MPEG These
`total of 15 documents
`documents together with the digital video tape for subjective testing and
`tape for
`computer
`further verification were requested of proposer according
`to the proposal package description
`Document MPEG89/ 128 adopted in the Stockholm meeting In the first day of the verification
`the 15 proposals were presented orallyby their proposers with
`particular attention given to the
`points that the proposers thought were novel ideas orjust improvements After the presentations
`two verification teams lead by Mr Haskell and Mr Watanabe were formed in order to verify the
`claims of the proposers in particular in matters concerning
`the amount of bits used to compress
`thetest sequence
`the figure of merit for random access
`the technical solutions for playing reverse
`the technical solutions for performing fast searches
`the compatibility with CCITr p64 and JPEG
`image mode
`the still
`
`Because of the amount of work required by the verification of 15 proposals the two teams
`operated in parallel for three half days At the same time those of the 15 proposals with VLSI
`study were verified by the VLSI Ad-Hoc Group
`liaison was established
`between
`the two
`verification teams and the VLSI group since many claims had also to be checked from the point of
`view of VlSI The verification teams reported on all the proposals
`and contributed
`the figure of
`merit for random access they also singled out some wealmesses of certain proposals in particular
`with regard to exceeding the target bit rate of 0.9 Mbit/second decided in Stockholm for this round
`of testing
`
`2/ Kurihama meeting Commonality and Differences
`
`between approaches
`
`proposals
`
`After taking some time for the members of MPEG-Video to review and analyse the results of
`the subjective testing and the verification the meeting focused on how to evolve the very promising
`results viewed and discussed during the week into standard Because there was feeling that many
`lot in common 13 out of 15 used 8x8 DC it was thought
`had
`that it was possible to
`Reference Model as
`common basis for improvement of the compression algorithm It
`turned out however
`useful Reference Model
`that
`is not
`superset of all the proposals
`but
`requires instead some hard choices and that the comittee had not sufficiently
`experimented with
`
`define
`
`9
`
`

`

`and
`to make those choices In order to further understand the commonality
`the possible options
`the differences between the many proposals Core Algorithm group composed of proposers was
`formed The discussion
`in the core algorithm group was centered around trying to establish the
`difference between what each proposer judged important and what
`they viewed as detail
`
`consensus on the use of DCT and motion compensation
`to be
`Because there appeared
`taxonomy of the proposals was proposed according to how the interframe redundancy is exploitedt
`could be classified into four classes
`According to this taxonomy the proposals
`
`A\ Standard Predictive coding
`CECAS DEC IBM Fujitsu NEC NiT Mitsubishi
`
`B\ Standard Predictive coding
`Interpolation
`Bellcore Matsushita Philips Telenorma/U Hannover
`
`C\ Non-causal Predictive coding with past and future reference
`C-cube JVC SONY
`
`D\ Non-causal Predictive coding with central
`AlT
`
`reference
`
`This classification
`
`of
`
`tool because within
`each class the task of
`useful
`the proposals
`reference model and
`program Of work is made easier Also many proposers using
`establishing
`declared themselves ready to try solutions including Interpolation
`Standard Predictive coding
`The solution in
`and
`hybrid between
`
`is
`
`is
`
`3/Osaka meeting
`
`path toward the standard
`
`During the Osaka meeting after discussing the Taxonomy established
`Group it was decided to form two sub-groups with the goal to
`
`by the Core Algorithm
`
`narrow down the choise of selection
`
`identify strong algorithm component
`mix best of diverse proposals
`
`Objective
`
`reference model The two groups were formed
`of the sub-groups is to achieve
`of the taxonomy with one group G.1 studying mainly uni-directional
`and
`around classes
`prediction in combination with motion compensated
`interpolation of every second frame while
`the other group G2 will
`past and future reference
`prediction based on
`focus on non-causal
`There is of course some overlap
`between
`the two frameworks since the interpolation of G.1 is
`special case of the non causal prediction of G.2 and both sub-groups are encouraged
`information with the other group
`
`to exchange
`
`The following companies
`have expressed the intention to work within Group
`AU Beilcore CECAS IBM Intel GCT Fujitsu Matsushita NiT Philips Siemens Sony
`Telenorma University of Hannover
`and Mr
`Mr
`Hepper Deutsche Thomson-Brandt will lead Group
`Kogure Matsushita
`Goizales IBM will act as regional coordinators
`for Japan and North-America
`and
`
`have expressed the intention to work within Group
`The following companies
`C-cube DEC NEC JVC Mitsubishi
`
`10
`
`

`

`Schedule
`
`for sub-groups
`
`1/ Proposers identity most promissing components
`head by November
`15th
`
`from their proposal and fax/mail
`
`to group
`
`2/ Group heads distribute information to all the sub-group members by November 28th
`
`3/ Group members make
`selection of components to implement/investigate
`enviroment and inform group heads by december 4th
`
`in their own
`
`4/Group heads make sure tlatallpromising components
`individual members for working additionally on them
`
`are investigated and if necessary contact
`
`5/Subgroup meet separately in January to arrive at one reference model for each
`
`6/Simulation resultsfor January may be provided by U-Matic tape in either 50 or 60 Hz format
`The target bit rate for simulationis
`1.15 Mbits/s
`
`The meeting of the sub-group is going to take placeJanuary 2323 and 25th at
`European location
`
`yet unknown
`
`4/ Pre/post Processing Group
`
`At some time during both the Kurihama and Osaka meeting the iusse of input
`formats coding
`formats and display format was ra sed At this point and while the input and output formats were
`fixed to the CCIR 601 422 fofmat in both the 525/60 and 625/50 variants most proposers use
`different coding format e.g 352x240 1130352288
`125 The rangeof solutions used to convert
`from the input
`format to the coding format vary widely from proposal to proposal Since pre and
`can be viewed as falling outside the domain of standardization but have
`post processing
`operations
`strong impact on the quality of the decoded
`video signal
`common
`it is important
`to establish
`so that the experts in MPEG-Video can focus theirwork on coding Also since
`pre/post
`processing
`the word has been using at least
`for moving pictures 2425 and 30Hz the issue
`three framerates
`of frame rate conversion is to be addressed In many uses of MPEG the display format is not going
`to be fixed and the issue of spatial
`format conversion also need to be address For all those reasons
`Special Group task force
`on pre/post
`and format conversion will operate under
`processing
`the leadership of Mr
`HaskelL The members of this group are individuals and more volunteers
`are welcome
`
`5/ Other Items/decisions
`
`During the Osaka meeting
`
`number of other points were discussed
`
`Bit rate the target bit rate for MPEG video is 1.15 Mbitx/s Quality at this rate is very important
`Number of Intra coded ftrzmes the number of intracoded
`be at least
`two
`
`frame per second access points must
`
`11
`
`

`

`Video tape ofproposats the video tape of the proposal tested at the Kurihama meeting will not
`be publicly available For the purposeof algorithmic work only the tape can be made available to
`the proposers under specific
`conditions
`
`demonstration tape of tile state of the art of coding within MPEG will be
`Demonstration tape
`made available For contributions
`contact MPEG-Video Chairman
`
`Annex
`Attendance List
`
`Asai
`
`Butera
`
`LChiariglione
`
`Fujiwara
`if Fukutome
`
`Gonzales
`Goto
`
`Guichard
`Hammer
`Haskell
`
`Hepper
`Hidaka
`Hontama
`
`Inoue
`
`Ishikawa
`
`Ishikawa
`
`Judice
`
`Kamikura
`
`Katayama
`Kogure
`Kummerow
`Le Gall
`
`Ligtenberg
`Matsuda
`
`Matsui
`
`Matsuyama
`Morris
`
`Morrison
`Murakami
`
`Nagata
`Normile
`
`Oestberg
`Ohta
`
`Okamoto
`Onoe
`
`Porter
`
`Pun
`
`Saint-Girons
`
`Sebestyen
`Simon
`
`Smith
`
`Sugiyama
`
`Sugiyama
`Szabo
`
`Takahashi
`
`Mitsubishi Electric Corp
`IT Intermetall
`CSELT
`GCT
`ASCII Corp
`International Business Machines Co
`Matsushita Electric Industrial Co Ltd
`CNET
`Siemens AG
`American Telephone
`and Telegraph
`Deutsche Thomson-Brandt GmbH
`Victor Company of Japan Ltd
`OKI Electric Co
`Matsushita Electric Co Ltd
`Ricoh Company Ltd
`Ricoh Company Ltd
`Bell Communications Research
`Nippon Telegraph and Telephone Corp
`GC1
`Matsushita Electric Industrial Co Ltd
`Telefonbau und Normalzeit GmbH
`Bell Communications Research
`C-Cube Microsystemsorp
`Fujitsj Limited
`CASIO Computer Co Ltd
`Yamaha Corp
`Philips Research
`British Telecommunications
`Mitsubishi Electric Corp
`Matsushita Electric Industrial Co Ltd
`Apple Computers
`SAFAD
`NEC Corp
`hitachi Ltd
`Victor Company of Japan Ltd
`Cypress Semiconductor
`American Telephone
`and Telegraph
`Thomson Consumer Electronics
`Siemens AG
`Intel Corp
`INMOS Ltd
`Toshiba Corp
`Victor Company of Japan Ltd
`Digital Equipment Corp
`Matsushita Electric Industrial Co Ltd
`
`Laboratories
`
`plc
`
`8146747 1251
`49761517 880
`39 11 2169520
`8134987543
`813 499 2146
`USA 1914 945 2141
`816 906 1748
`33 14529 5435
`4989636 48322
`12019493697
`495114182483
`81468368540
`8137987631
`8144934 2862
`81455933489
`
`USA
`
`USA
`
`81455933489
`12018297019
`8146959 2829
`8134987543
`8169061748
`49692723315
`USA
`12018297019
`USA 1415496 2205
`8144754 2582
`81425632456
`
`GB
`GB
`
`44293776495
`44473643791
`81467 47 1251
`8169061748
`
`USA
`
`408 974 1794a
`468 137025
`8144856 2239
`8145 864 4305
`8146836 8844
`USA 1408943 2949
`USA
`12019493697
`USA 1213568 9002
`498972247713
`
`USA
`GB
`
`USA
`
`609 936 7700
`44454617910
`81445492261
`8129735605
`15084867417
`816 906 1748
`
`12
`
`

`

`Takeguchi
`Tanaka
`
`Texier
`Tsuda
`Van der Meer
`Watanabe
`
`Wright
`Yamada
`KYang
`Yasuda
`
`Yasura
`CYeh
`Yonemitsu
`Yung
`
`Institute
`
`Matsushita Electric Industrial Co Ltd
`Matsushita Research
`CNFF
`Fujitsu Laboratories Ltd
`Philips Consumer Electronics
`Nippon Telegraph and Telephone Corp
`Apple Computers
`Victor Company of Japan Ltd
`Bell Communications Research
`Nippon Telegraph and Telephone Corp
`Victor Company of Japan Ltd
`Apple Computers
`Sony Corp
`Apple Computers
`
`NL
`
`USA
`
`USA
`
`8169061748
`81449331797
`33145295435
`4-81447542582
`3140758503
`81468592829
`14089746615
`81468368540
`12018297019
`81468592306
`81468368540
`14089746615
`81462306284
`USA 14089746615
`
`USA
`
`13
`
`

`

`Annex
`
`Separate Meeting of the MPEG/Audio Group
`
`Source
`
`Status
`
`Hans-Georg Musmann
`Chairman of MPEG/Audio
`Draft
`
`During the three days from October
`
`18 to October20 the MPEG/Audio group discussed audio
`related problems in separate sessions at the Kuriharna RD Center Victor Company of Japan
`JVC in Yokosuka/Japan The list of participants of these sessions is given in Annex
`
`detailed description of the algorithm and an indication Ofthe intention
`By this October meeting
`groups has been requested for each of
`to join one of the proposed development
`the 14 pre
`registered audio coding algorithms 12 of the 14 algorithm proposers followed the encouragement
`the chairman and built the following
`groups with two remaining individual
`of
`development
`
`algorithms
`
`Transform Coding with overlapping
`ATT DTB FhG development group
`CNET individual
`
`blocks
`
`Transform Coding with non-overlapping blocks
`Fujitsu JVC NEC Sony
`development group
`
`Subband Coding with less than or equal to
`BT NiT
`development group
`
`subbands
`
`Subband Coding with more than
`subbands
`CCE1TT IRT Philips development group
`Matsushita
`individual
`
`candidates CNET and Matsushita
`The remaining two individual
`explained that
`interval since the Stockholm meeting had been too short
`for successful negotiations
`but both
`indicated their willingness in principle to join also the adequate development group With respect
`new deadline
`to the chance of avoiding
`time consuming
`preselection tests the group agreed on
`1989 Both CNEF and Matsushita were strongly encouraged
`for clustering Friday December
`by the chairman to join one of the
`development groups by this deadline
`
`the time
`
`group algorithms left no preselection is necessary Otherwise
`In case that there are only
`to be clarified whether
`the final
`tests could be performed with
`algorithms or preselection
`or
`test has to be carried out
`
`it has
`
`one for each of the four development groups and the two
`Six detailed algorithm descriptions
`individual proposals had been submitted and registered as MPEG documents
`89205 through
`the amount of detail given in these six descriptions
`891210 on the first day of
`the meeting Since
`differed more or less the audio group agreed on the following procedure
`
`Document MPEG 89/205
`has been chosen as
`guideline for the amount bf detail
`to be given
`by the detailed description of the algorithm Furthermore the detailed description has to con
`tain encoder and decoder
`
`14
`
`

`

`Every algorithm proposer should
`that it matches the guideline
`
`if necessary
`
`revise his detailed description in such way
`
`The following deadline
`for deliveiy of the revised detailed descriptions to the MPEG/Audio
`chairmans office has been fixed Friday November 17 1989 That
`is the revised detailed
`descriptions of algorithms are requested to arrive at latest on November 17 1989 at the office of
`Prof Musniann
`
`In case that the remaining two individual proposals
`decide to join one of the
`development
`groups the detailed description of the newly compromised algorithm of the relative groups has
`to be submitted at the next meeting of the MPEG/Audio group which is planned for December
`11 and 12 1989
`
`The chairman reminded all participants
`that
`the distribution of the detailed descriptions of
`algorithms is restricted to the MPEG/Audio group and therefore nobody is allowed to distribute
`outside the MPEG/Audio group
`these documents
`
`arranging
`
`registered
`
`The EBU European Broadcasting Union informed the chairman about
`its serious interest in
`the subjective testing of the audio systems under investigation inthe MPEG/Audio group
`see document MPEG 89/199
`first company the Swedish Broadcasting Corporation Sveriges
`Radio SR indicated its offer also directly to the chairman and
`discussion took place
`preliminary
`just before this meeting The chairman informed the group about
`these activities
`and
`short
`discussion resulted in an unanimous acceptance of this offer Mr Deheiy CCETF informed the
`group that there are one or two other possible candidates within the EBU and during the meeting
`second offer by Thames Television UK No other volunteer has been
`the chairman received
`It was agreed to continue the negotiations with SR and to keep Thames Television as
`an alternative An intermediate MPEG/Audio meeting was scheduled for December
`11 and 12
`ifpossible SR Stockholm in order to give the experts ofSR the opportunity
`1989 at
`to participate
`in the detailed discussions
`the test procedures and the organization of the tests An
`concerning this intermediate meeting will be distributed
`as soon as
`
`invitation with all details
`
`regarding
`
`possible
`
`Most of the time during this meeting was spent for refining the Proposal Package Description
`Mainly the group agreed on framework for the subjective testing and scoring of the performances
`to be tested
`an extension
`of the requirement
`list and
`refinement of the time schedule towards
`DP The results of these discussions
`can be found in the respective paragraphs of the newest draft
`Document MPEG 89/128
`version of the Proposal Package Description
`
`The number of audio-related implications
`to be considered for testing has been reduced again
`so that there are 11 remaining performances to be tested
`by
`framework for subjective testing
`and scoring of these 11 performances
`has been agreed on The decision about some details is
`postponed to the December meeting which hopefully could be hosted by the company which will
`carry out the final subjective testing in order to exploit the experiences
`and the special knowledge
`of its experts The details of the agreed framework
`can be found in paragraph 72 Performances
`to be tested sound and paragraph 82 Deflnition of test procedure sound of the PPD Doc
`MPEG 89/128
`
`strong request of the MPEG chairman Dr Chiariglione
`According to
`the performance toll
`quality was discussed again As agreed in Stockholm this implication will not be tested but due
`to the importance of toll quality for several possible applications
`the group agreed on the
`following procedure
`
`15
`
`

`

`Eveiy proposer orgroup ofproposei shou