Compression video overview

Compression video overview

演講者：林崇元

Outline Introduction Fundamentals of video

compression Picture type Signal quality measure Video encoder and decoder Standard’s

Introduction Why we need to compression

Picture A picture consists of three rectangular matrices

representing luminance (Y) and two chrominance (Cb and Cr) values

The Y matrix has an even number of rows and columns

The Cb and Cr matrices are one-half the size of the Y matrix in each direction (horizontal and vertical).

Introduction Applications for image, video, and audio compression

Introduction Achieve high compression performance

while keep good picture quality Theorem

Spatial redundancy – DCT,DFT,subband,wavelet

Temporal redundancy – MC/ME Statistical redundancy – VLC, Entropy

coding Perceptual redundancy – VQ

Introduction Tradeoffs in lossy compression

Fundamentals of video compression Use the technique of the JPEG

DCT based coding scheme DCT transform (2D)

Fundamentals of video compression Use the technique of the JPEG

Discrete cosine transform

Fundamentals of video compression Use the technique of the JPEG

DCT based coding system

Image

Spatial-to-DCT domain transformation

8 x 8 DCT

Lossless coding of DCT domain samples

Entropy Coding

Discard unimportant DCT domain samples

Quantization

Fundamentals of video compression Quantization

Eyes are insensible to high-frequency components

The greater quantizer means greater loss

Lower frequency component has smaller quantizer, high frequency component has greater quantizer

The quantiation tables in the encoder and decoder are the same

Fundamentals of video compression Use the technique of the JPEG

The spatial domain is redundancy For the DCT-based coding system on an

image-by-image, one can achieve close to 14Mbits per second, which is too high for practical uses

For lower bit rate, we must introduce temporal redundancy

Fundamentals of video compression Temporal redundancy

The temporal correlation in an image sequence

Fundamentals of video compression Temporal redundancy

Instead of 3-D DCT, most video coders use a two-stage process to achieve good compression

Two-stage video coding process

Fundamentals of video compression Temporal redundancy

Motion estimation

Fundamentals of video compression Temporal redundancy

Full search algorithm

Picture type Video bit stream

Picture type Slice

One or more "contiguous'' macroblocks. The order of the macroblocks within a slice is from left-to-right and top-to-bottom.

Macroblock A 16-pixel by 16-line section of luminance components

and the corresponding 8-pixel by 8-line section of the two chrominance components.

Block A block is an 8-pixel by 8-line set of values of a

luminance or a chrominance component.

Picture type Intra picture

Coded using only information present in the picture itself

I-pictures provide potential random access points into the compressed video data.

I-pictures use only transform coding

Picture type Predicted picture

coded with respect to the nearest previous I- or P-picture.

P-pictures use motion compensation Unlike I-pictures, P-pictures can propagate

coding errors

Picture type Bidirectional picture

Coded use both a past and future picture as a reference

B-pictures provide the most compression and do not propagate errors

Picture type The choice of picture type

The MPEG algorithm allows the encoder to choose the frequency and location of I-pictures is based on the application's need for random accessibility and the location of scene cuts in the video sequence

The encoder also chooses the number of B-pictures between any pair of reference (I- or P-) pictures. This choice is based on factors such as the amount of memory in the encoder and the characteristics of the material being coded

Picture type Typical display order of picture types

Video stream composition The MPEG encoder reorders pictures in the video

stream to present the pictures to the decoder in the most efficient sequence

Signal quality measure SNR (signal-to-noise ratio)

encoder input signal energy

SNR = 10log10

noise signal energy

PSNR (peak signal-to-noise ratio) Instead of using the encoder input signal,

one uses a hypothetical signal with a signal strength of 255

Video encoder

Video decoder

MPEG-1 Media storage Optimal for frame size 352x240x30 Bitrate : up to 1.5 Mbit/s International standard in 1992

MPEG-2 Applications from storage to HDTV Bitrate : standard definition TV:4-9 Mbit/s HDTV:15-25 Mbit/s Interlaced/non-interlaced Scalability Capable of decoding MPEG-1 bitstream International standard in 1994 Single chip for video and audio

MPEG-4 Applications for multimedia communication Bitrate : 10K-25 Mbit/s Object – based coding Natural and synthetic video Scalability Robust and error resilience International standard in 1998 Single chip for video and audio