Image Processing Reference
The MPEG-1 Codec
The MPEG-1 codec is still used for some legacy applications but the capabilities of MPEG-
2 and MPEG-4 provide better compression ratios and more sophisticated coding tech-
niques. New projects are unlikely to be based on MPEG-1. This was a popular codec when
deploying video on CD-ROMs and is also used in games design.
How It Is Done in MPEG-1
The simplest of the MPEG video-compression formats is the best place to start. The others
are all based on this model and just add some enhancements. The inventors of the MPEG
coding schemes considered concepts of frames and luma plus chroma representation and
worked out some techniques for compressing video that provide the most flexible but
The MPEG-2 and H.264 (MPEG-4) codecs are similar in many respects and will be
easier to explain once MPEG-1 compression is understood.
The MPEG-1 standard was constructed around the limitation that the bit stream
must be playable on a CD-ROM having a bit rate of about 1.4 Mbps. This must include
compressed audio and video in a bit stream that previously was designed to deliver only
digital audio content when CDs were first invented. This presents a significant challenge.
If you already know a little bit about how JPEG compression works, then you will
understand the core of this topic. MPEG modifies the JPEG encoder slightly and adds
capabilities to deal with sequences of frames containing motion pictures.
So let's now dig down deeper into the MPEG-1 compression system.
The output of an encoder is a bit stream. The standards describe the syntax of this bit
stream and the encoder must produce an output that is compliant. The syntax is structured