What is H.264 / MPEG-4 AVC?
H.264 is one of the standards for video data compression encoding recommended by the ITU (International Telecommunications Union).
It is also recommended by ISO (International Organization for Standardization) as part of the MPEG-4 video compression standard (MPEG-4 Part 10 Advanced Video Coding). For this reason, in general, it is often called simply H.264, or both names are written together, such as “H.264 / MPEG-4 AVC” and “H.264 / AVC”.
H.264 is said to be able to compress the size to about half that of MPEG-2 due to its high data compression rate, which is said to be more than double that of MPEG-2, which has been widely used in the past. , and more than 1.5 times that of MPEG-4.
Sending the modified part of the image is the same as with other compression methods, but in H.264, the whole screen is divided into smaller parts and only the modified information is sent, so the amount of information is reduced even for the same modified image. do.
Also, regarding the modified information, with MPEG-2, the modified information of the image can be acquired only in units of 0.5 pixels, so there will be a difference from the previous screen. Since the difference information needs to be sent, the amount of information sent at one time is large. On the other hand, with H.264, the change information can be acquired in units of 0.25 pixels, so there is no difference between the screen after receiving the change information and the target screen. Therefore, the amount of information sent at one time is small.
For these reasons, H.264 is used for low-speed and low-quality applications such as videophones for mobile phones, video conferencing systems, One-Seg terminals (digital terrestrial transmission service for mobile phones), games, and next-generation optical devices. . The “HD-DVD, Blu” discs. -Ray Disc “are used in a wide range of applications, including high-quality, high-capacity video such as HDTV broadcasting.
The basic H.264 encoding method follows the conventional method, such as H.263, and uses an algorithm that combines motion compensation, inter-frame prediction, DCT (discrete cosine transform), entropy coding, etc. For each technology, we have improved the processing method, such as replacing floating point arithmetic with integer arithmetic, and by incorporating new technology, we have achieved a better compression ratio than the conventional method.
Since there are several methods to choose from regarding frame prediction technology and encoding, the multiple combinations of these are defined as “profiles”. By using it correctly in accordance with the purpose, it is possible to respond flexibly to the difference in required processing performance and bit rate
