written 8.4 years ago by | • modified 4.5 years ago |
written 8.4 years ago by | • modified 8.4 years ago |
Main features of Video Compression:
- Subsampling
- Block based coding
- Difference Coding
MPEG Industry Standard for Video Standard
MPEG 1 Video Standard
1.Blocks (8×8) in size of either an original from or difference between a (frame) and (motion compensated prediction) are transformed using DCT.
2.The transformed coefficients are quantized and transmitted to the receiver.
3.A buffer is used for smooth delivery of bits from the encoder and also for rate control.
4.The major contribution of MPEG 1 is “ Random Access Capability” This is done by inserting I frames frequently. However compression ratio is low for I frames. The efficiency increases because MPEG 1 is also contains 2 other types of frames P (Predictive Coded) and B (Bidirectional Predictive Coded) frames.
5.The different frames are organized together in a “group of pictures” (GOP). GOP has to contain at least I frame.
6.There are 2 different sequence orders
- Display Order: It is the sequence in which the video sequence is displayed to the user.
- Processing order in Bit Stream Order: It is the order in which the frames should be decoded.
MPEG 1 parameters are known as ‘Constrained Parameter Bitstream” (CPB)
- Horizontal Picture Size ≤ 768 pixels
- Vertical Picture Size ≤ 576 pixels
- Pixel rate = 396 blocks / frame of frame rate 25 frames/ sec
- Video Compression = 330 bits / frame if fr = 30 frames /sec
MPEG 1 | Data rate is 1.5 Mbps |
---|---|
MPEG 2 | Data rate 10 Mbps ≤ |
MPEG 3 | HDTV compression was merged with MPEG 2 |
MPEG 4 | Low data rate < 64kbps |
7.MPEG-2 Video Standard
- To provide generic application, independent standard.
- It uses “Tool Kit Approach”
- For a particular application user can select any profile or level from a set of profiles and levels
- Profiles- Define the algorithms to be used.
- Levels- Define the constraints on the parameter.
- 5 Profiles -1. Simple, 2. Main, 3. SNR scalable, 4. Spatially Scalable, 5. High
- Each higher profile is capable of decoding video encoded using all profiles upto and including that profile.
- Decoder designed for ‘snr-scalable’ could decode value that is encoded using simple, main, snr scalable.
- Simple Profile