Video Compression is the science of reducing transmitted video data without noticeable loss of quality to the original image recorded. It enables transmission using reduced bandwidth for satellite, terrestrial, modem, DSL and ISDN broadcasts. Compression is also used to store digital video. Compression results from an analysis of the original video data source which decides which data to keep and which can be discarded. Discarded data is information that cannot easily be seen by the human eye.
Common Code Analysis
The analysis gives each event a code depending upon its frequency; common occurring events are coded with few bits and less common events are coded with more bits. This encoding is known as signal analysis, quantization and variable length encoding. Too much compression can result in loss of quality.
Compression Video Technology
There are four methods of compression; Discrete Cosine Transform (DCT), Vector Quantization (VQ), Fractal Compression (FC), and Discrete Wavelet Transform (DWT). Most of the compression methods are known as Lossy Compression - this is where files are permanently reduced by discarding irrelevant information. When the file is uncompressed, the data is not the same as the original file.
Understanding Compression Standards
MPEG (Moving Image Picture Experts Group)
Founded in 1988 to develop standards for digital audio and video formats, five MPEG standards have now been developed. The different compression standards were designed for specific applications and bit rates. These are:
MPEG1
(Compression rates up to 1.5 Mbit/sec)
The compression standard of moving pictures with audio. This standard was designed for CD-ROM video applications. It is a widely-used standard for Internet videos which are transmitted as .mpg files. VideoCD uses the compression standard MPEG-1 which is used extensively throughout Asia for video distribution. MP3 is the compression standard MPEG-1 Level 3 and is the most widely used standard for digital compression of audio.
MPEG-2
(Compression rates between 1.5 and 15 Mbit/sec)
This is the standard upon which digital television set top decoders and DVD compression is based. This compression standard is based on MPEG-1, but was designed specifically for digital television broadcast. The main improvement over MPEG-1 is its ability to compress interlaced video efficiently. There was never any requirement to develop MPEG-3 compression as MPEG-2 adapts well to HDTV bit rates and resolution.
MPEG-4
This standard was developed for multimedia and Web compression. MPEG-4 is used for object-related compression rather like Virtual Reality Modelling Language. MPEG-4 files are created by compressing together individual objects within a scene that have been tracked separately. This is an efficient, scalable compression from low bit rates through to very high. It enables the control of objects independently within a scene, introducing interactivity.
MPEG-7
This standard, currently under development, is also called the Multimedia Content Description Interface. When released, the group hopes the standard will provide a framework for multimedia content that will include information on content manipulation, filtering and personalization, as well as the integrity and security of the content. Contrary to the previous MPEG standards, which described actual content, MPEG-7 will represent information about the content.
MPEG-21
Also known as Multimedia Framework, MPEG-21 has been designed to describe the elements required to compile a framework for the delivery and consumption of multimedia content, and their relationship within.
JPEG (Joint Photographic Experts Group)
A working party that develops standards for continuous tone image coding, JPEG is a lossy compression format which is used for colour or grey-scale pictures. It works by making use of the fact that the human eye does not notice small colour variations. JPEG 2000 is an image coding format which uses a compression method based on the of wavelet technology.
DV
Digital video is a lossy compression standard providing a high-resolution format for video cameras and camcorders. It uses discrete cosine transform (DCT) to compress the pixel data. Video captured using this format maybe transferred from the Camera/Camcorder using FireWire (IEEE 1394), which is a serial bus able to transfer data up to 50 MB/sec.
H.261
An ITU standard designed for video conferencing as a means of two-way communication via ISDN lines. Data rates are supported in multiples of 64Kbit/s. The algorithm is based on discrete cosine transform (DCT) which can be produced in hardware or software using intraframe and interframe compression, with resolutions supported by H.261 are CIF and QCIF.
H.263
Based on H.261 and has been enhanced to give improved video quality for modem transmission, resolutions supported are CIF, QCIF, SQCIF, 4CIF and 16CIF.
DivX Compression
DivX is a software application to enable video to be downloaded through a DSL/Modem connection. MPEG-4 is the standard used to compress digital video.