Video and Image Processing Blockset 2.8

Video Compression

This demo illustrates video compression using motion compensation and discrete cosine transform (DCT) techniques with the Video and Image Processing Blockset™. The demo calculates motion vectors between successive frames and uses them to reduce redundant information. Then it divides each frame into submatrices and applies the discrete cosine transform to each submatrix. Finally, the demo applies a quantization technique to achieve further compression. The Decoder subsystem performs the inverse process to recover the original video.

Contents

Demo Model

The following figure shows the Video Compression model:

Encoder Subsystem

The Block Processing block sends 16-by-16 submatrices of each video frame to the Block Processing block's subsystem for processing. Within this subsystem, the model applies a motion compensation technique and the DCT to the video stream. By discarding many high-frequency coefficients in the DCT output, the demo reduces the bit rate of the input video.

Video Compression Results

The Decoded window shows the compressed video stream. You can see that the compressed video is not as clear as the original video, shown in the Original window, but it still contains many of its features.

Available Demo Versions

Intensity version of this demo:

vipcodec.mdl

Color version of this demo:

vipcodec_color.mdl