Page 69 - Video Coding for Mobile Communications Efficiency, Complexity, and Resilience
P. 69

46                                      Chapter 3.  Video Coding:  Standards


            designed  as  a  toolbox,  where  the  user  can  decide  which  tools  to  use  for  the
            particular  application.  In  addition,  the  standard  de/nes  only  the  decoder  and
            the bitstream syntax. This allows a large degree of freedom for manufacturers
            to propose their own optimized encoders. This generic design and large degree
            of  freedom  have  contributed  to  the  success  of  MPEG-1.  It  has  been  used  in
            a  wide  range  of  applications,  from  interactive  systems  on  CD-ROM  to  the
            delivery of  video over  telecommunication networks.

            3.3.6  MPEG-2

            In  1990,  ISO=IEC  JTC1=SC29=WG11  started  studies  on  a  new  standard  for
            applications  not  covered  by  MPEG-1.  In  particular,  the  new  standard  was
            intended to provide video quality not lower than NTSC=PAL and up to CCIR-
            601 quality at rates around 10 Mbits=s. This standardization activity was nick-
            named  MPEG-2  because  it  was  seen  as  phase  2  of  the  work  started  in
            MPEG-1.  In  1992,  ITU-T=SG  15  joined  this  standardization  e1ort  to  develop
            video  coding  for  Asynchronous  Transfer  Mode  (ATM)  networks.  In  1993,
            it  was  realized  that  the  scope  of  MPEG-2  could  be  enlarged  to  suit  cod-
            ing  of  HDTV.  This  made  an  initially  planned  MPEG-3  for  HDTV  superJu-
            ous.  In  1994,  the  ISO=IEC  13818  (MPEG-2)  standard  (ITU-T  Recommen-
            dation  H.262)  was  drafted  [63],  and  later  in  the  year  it  was  accepted  as  an
            international  standard.  Like  MPEG-1,  the  MPEG-2  standard  is  generic  and
            Jexible.  In  fact,  MPEG-2  can  be  thought  of  as  a  superset  of,  and  as  such
            was  designed  to  be  backward  compatible  with,  MPEG-1.  There  are  many
            additional  features  provided  by  MPEG-2  over  MPEG-1,  including  the  sup-
            port  for  interlaced  video  and  scalability.  Since  implementation  of  the  full
            MPEG-2  syntax  may  not  be  practical  for  most  applications,  MPEG-2  has  in-
            troduced  the  concepts  of  “pro/les,”  describing  functionalities,  and  “levels,”
            describing  resolutions,  to  provide  subset  conformance  levels.  MPEG-2  has
            had  even  more  success  than  MPEG-1,  with  applications  in  the  areas  of  ca-
            ble  TV,  networked  ATM  services,  and  satellite  and  terrestrial  TV  broad-
            casting.


            3.3.7  H.263
            The  increasing  demand  for  digital  video  communications  over  the  public
            switched  telephone  network  (PSTN)  and  mobile  networks  initiated  a  new
            standardization e1ort by ITU-T=SG 15. The aim was to develop a video cod-
            ing  standard  for  low-bit-rate  applications  below  64 kbits=s.  The  result  of  this
            e1ort was ITU-T Recommendation H.263 [64], which was completed in 1995
            and  approved  in  1996.  Although  H.263  was  based  on  the  coding  structure
            of  H.261,  it  provides  a  signi/cant  improvement  in  performance.  Side-by-side
   64   65   66   67   68   69   70   71   72   73   74