MPEG stands for Moving Picture Experts Group. This name originally referred to an organization that studies video and audio encoding standards, established in 1988, dedicated to developing compression coding technologies for video and audio. Now, when we refer to MPEG, we mean a series of video encoding standards set by this group. The group was formed in 1988 and has since established several standards including MPEG-1, MPEG-2, MPEG-3, MPEG-4, and MPEG-7, with MPEG-21 currently under development. MPEG is a working group of ISO and IEC, with its official title being: First Technical Committee, Subcommittee 29, Working Group 11, in English as ISO/IEC JTC1/SC29 WG11. MPEG holds meetings approximately every 2-3 months, each lasting about 5 days. During these meetings, new proposals and technical details are discussed within the group before moving into the formal standardization audit process. Other video standardization working groups related to the MPEG working group include ITU-T VCEG and JVT.
To date, MPEG has developed and is developing the following video-related standards:
MPEG (usually referring to MPEG-1) video encoding is based on transform-based lossy compression. Optical signal lines are sampled to form video signals, where the basic unit of the video signal is called a frame, and a frame is an independent image. The frame is then divided into small blocks for transform coding, followed by quantization, and finally entropy coding.
MPEG-1, MPEG-2, and MPEG-4 actually utilize motion estimation and motion compensation techniques. In frames (images) using motion compensation, what is encoded is the difference between the reference frame compensated for motion and the current image. Unlike traditional image encoding techniques, MPEG does not compress every image frame but instead compares each frame within a one-second period. Since typical video content usually has little background change and significant subject change, MPEG technology applies this characteristic by treating one image as the main image while recording only reference and change data for the remaining frames, effectively recording dynamic images. From MPEG-1 to MPEG-4, the core technology still adheres to this principle, with differences mainly in the comparison process and complexity of analysis.
MPEG only specifies the bitstream format and decoding accuracy (i.e., the method of decoding), and anyone can implement encoders (programs) in different ways according to the MPEG standard. In addition to reducing commercial disputes arising from encoding patents, the main purpose of the MPEG standard is to ensure that bitstreams produced by different encoders can be correctly decoded by other decoders, as long as the bitstream conforms to the standard.
蘊藏許多助人的知識與智慧。