(Moving Pictures Experts Group) A family of ISO/ITU standards for compressing digital video. Pronounced "em-peg," it is the universal standard for digital terrestrial, cable and satellite TV, DVDs and digital video recorders (DVRs).
MPEG uses lossy compression within each frame similar to JPEG, which means pixels from the original images are permanently discarded. It also uses interframe coding, which further compresses the data by encoding only the differences between periodic frames (see interframe coding). MPEG performs the actual compression using the discrete cosine transform (DCT) method (see DCT).
MPEG is an asymmetrical system. It takes longer to compress the video than it does to decompress it in the DVD player, PC, set-top box or digital TV set. As a result, in the early days, compression was perfomed only in the studio. As chips advanced and became less costly, they enabled digital video recorders, such as Tivos, to convert analog TV to MPEG and record it on disk in real time (see DVR).
MPEG-1 (Video CDs)
Although MPEG-1 supports higher resolutions, it is typically coded at 352x240 x 30fps (NTSC) or 352x288 x 25fps (PAL/SECAM). Full 704x480 and 704x576 frames (BT.601) were scaled down for encoding and scaled up for playback. MPEG-1 uses the YCbCr color space with 4:2:0 sampling, but did not provide a standard way of handling interlaced video. Data rates were limited to 1.8 Mbps, but often exceeded. See chroma subsampling.
MPEG-2 (DVD, Digital TV)
MPEG-2 provides broadcast quality video with resolutions up to 1920x1080. It supports a variety of audio/video formats, including legacy TV, HDTV and five channel surround sound. MPEG-2 uses the YCbCr color space with 4:2:0, 4:2:2 and 4:4:4 sampling and supports interlaced video. Data rates are from 1.5 to 60 Mbps. See chroma subsampling.
MPEG-4 (All Inclusive and Interactive)
MPEG-4 is an extremely comprehensive system for multimedia representation and distribution. Based on a variation of Apple's QuickTime container format, MPEG-4 offers a variety of compression options, including low-bandwidth formats for transmitting to wireless devices as well as high bandwidth for studio processing. The widely used H.264 standard is based on MPEG-4. See H.264 and QuickTime.
MPEG-4 also incorporates AAC, which is a high-quality audio encoder. MPEG-4 AAC is widely used as an audio-only format (see AAC).
A major feature of MPEG-4 is its ability to identify and deal with separate audio and video objects in the frame, which allows individual elements to be compressed more efficiently. User-controlled interactive sequences that include audio, video, text, 2D and 3D objects and animations are all part of the MPEG-4 framework. For more information, visit the MPEG Industry Forum at www.mpegif.org.
MPEG-7 is about describing multimedia objects and has nothing to do with compression. It provides a library of core description tools and an XML-based Description Definition Language (DDL) for extending the library with additional multimedia objects. Color, texture, shape and motion are examples of characteristics defined by MPEG-7.
MPEG-21 (Digital Rights Infrastructure)
MPEG-21 provides a comprehensive framework for storing, searching, accessing and protecting the copyrights of multimedia assets. It was designed to provide a standard for digital rights management as well as interoperability. MPEG-21 uses the "Digital Item" as a descriptor for all multimedia objects. Like MPEG-7, it does not deal with compression methods.
The Missing Numbers
MPEG-3 was abandoned after initial development because MPEG-2 was considered sufficient. Because MPEG-7 does not deal with compression, it was felt a higher number was needed to distance it from MPEG-4. MPEG-21 was coined for the 21st century. See MP3, M-JPEG, MPEG LA, MPEGIF, Pro-MPEG Forum, MPEG-DASH, JPEG and interframe coding.