For as long as the factual production community has seriously dabbled in digital content (i.e., since it partied like it was 1999), MPEG2 has been the format of choice. Set-top boxes and DVD both use MPEG2, most archives that have moved into the digital age have adopted the format, and it’s the standard widely used for distribution of satellite television content. But, the rise of MPEG4 threatens to overthrow the reigning format. A sleeker, smarter standard, MPEG4 promises to speed things up, shrink things down, and present fresh possibilities for interactive content.
The Moving Picture Experts Group (MPEG), an international organization, first set the MPEG4 standards in October 1998 (Part 1), and expanded them in December 1999 (Part 2). ‘MPEG4 is an extremely broad standard compared to MPEG2,’ explains Randy May, consultant for the Cupertino, U.S.-based PixelTools.
‘[It offers] a whole range of new tools, not only for compression, but ones that can support interactivity.’
PixelTools itself works in the area of data compression, and this is where MPEG4 makes one of its huge leaps forward. After recently adopting what had been a parallel technology standard – a video-coding algorithm called H264 – MPEG4 now produces picture quality equal to or better than MPEG2, using three to four times fewer bits. By identifying only the pixels that change from frame to frame, and only coding the changes, the number of bits required for compression is dramatically reduced.
The advantages of the reduced bit-rate for a picture of equal or better quality are numerous: faster transmission, transmission of more channels across the same bandwidth, higher quality video on DVDs and the Web, and reduced storage requirements for editing and archives, to name a few of them.
What’s more, previously recorded MPEG2 digital video files can be updated and manipulated with the MPEG4 standard. May says this is crucial to the new standard. ‘This is a very important application,’ he affirms. ‘The focus is going to be on MPEG2, because that’s where so much of the legacy material is coming from.’
The second main thrust of the standard is what May calls ‘composition’ and what MPEG4 calls ‘the coding of audio-visual objects.’ Essentially, the software can encode, and therefore recognize, particular objects of arbitrary shape within the video stream – part of MPEG4′s ‘object-oriented approach.’
This capability is particularly exciting to people who are working in the area of interactivity. The isolation of objects in the frame will allow information to be tagged to specific parts of a frame, or cross-referenced to other parts of a film or other films altogether. It could allow the viewer to identify certain individual elements of a video (people or objects, for instance), and seek more information on those specific elements, or even choose to focus on that particular element or follow it rather than others. In effect, it could provide the viewer with options about how a story is told to them.
But, since technology to both generate and receive these kinds of ‘divided’ images is still being developed (and MPEG4 Part 10, Advanced Video Coding, is still being worked on), the full application of this aspect of MPEG4 remains somewhere in the future. How far? ‘I wish I knew,’ says May. ‘It would make business decisions a whole lot easier.’