
Types of videos

MPEG-1

The MPEG-1 video compression format was developed by the Motion Picture Expert Group, an international organization that creates video compression standards. It supports a maximum frame resolution of 4095 x 4095 pixels at a frame rate of up to 60 times per second. However, a resolution of 352 x 288 is generally used, which corresponds to the quality of recording on a normal VHS cassette.
How is the information compressed in this format? Suppose we have the following scene: the car moves from point “A” to point “B”. The movement of the machine can be described by two parameters: the vector of movement from point “A” to point “B” and the angle of rotation around its axis. At the same time, the background remains unchanged or almost unchanged: the viewer is unlikely to pay attention to the vibrations of small branches near distant trees. So you can divide the frame into two components: the background, which is saved once and then replaced during all-frame playback, and the area where the car moves, will have to be recorded separately for each frame.
Digital video: an overview of formats
In MPEG-1 format, all frames in a movie are classified into three types: I, P, and B frames. The first type (I-frames, Intra Frames) includes anchor frames. Your images are saved in their entirety in JPEG format. For the predicted frames, only the differences from the previous i-frame are recorded, which requires much less disk space. For B-frames (bidirectionally interpolated frames), the differences with the previous and next I or P-frames are preserved (Fig. 2).
Digital video: an overview of formats
As a result, the compressed file size is approximately 1/35 of the original. This means that an hour and a half movie with the quality equivalent to an analog VHS cassette in MPEG-1 format will fit on two CDs. For transmission over the Internet or in satellite transmission networks, this standard, of course, is not suitable. But there have been many consumer video CD players that work in this format (and they are still being produced now, by the way).
MPEG-2
MPEG-2 is an additional extension of MPEG-1. The recommended frame size has been increased; it is now 1920 x x 1080 pixels, added support for six channel audio. However, playing video in this format requires more processing power from your computer.
It should be noted that work was underway to create the MPEG-3 standard (not to be confused with the now popular audio compression format: MPEG-1 Audio Layer 3). It was supposed to become the basis for HDTV digital television systems. But the work was interrupted, because the necessary requirements for HDTV were implemented in the form of small extensions of MPEG-2.
This format is now quite widespread in the West: it is used to transmit video through satellite channels and digital cable networks, in addition, all DVD video discs are recorded in this format.
MJPEG
In fact, MJPEG (Motion JPEG) is a transition format from compressing normal photos to compressing videos. Each frame is recorded in JPEG format and then placed in the footage. MJPEG is mainly used in video editing cards like Fast AV Master, MiroVideo DC50, etc. You can reduce the video stream from 30MBps to 6MBps. For use in home video players, this standard is not very suitable due to the low compression ratio (5: 1) and the lack of means to synchronize video and sound.
MPEG-4
The MPEG-1 and MPEG-2 formats did not provide a real possibility to transmit video over the Internet and create interactive television based on them: the file size was too large. To drastically reduce it, as well as implement other functions necessary for video transmission, work has begun on the specifications of a new format: MPEG-4. In fact, it does not focus so much on video compression as on the creation of so-called “multimedia content”: the fusion of interactive television, 3D graphics, text, and so on.
As for the video itself, the most important innovation was the further improvement of technologies for decomposing a scene into objects and algorithms for their effective compression. Thus, for example, when compressing a video recording of a tennis match, most of the frames can be conditionally decomposed into images of the stands (background) and the players.







