Adaptive compression of animated sequences

Author

McNeill, D.K. ; Card, H.C.

Author_Institution

Dept. of Electr. & Comput. Eng., Manitoba Univ., Winnipeg, Man., Canada

Volume

1

fYear

1995

fDate

15-16 May 1995

Firstpage

191

Abstract

Unsupervised artificial neural learning algorithms have clearly demonstrated their practical value for vector quantization of speech and stationary images. We examine competitive learning as an adaptive compression algorithm for highly repetitious video signals such as those found in animated cartoons. The characteristics of animated video are presented and their suitability for compression are examined. An artificial neural network system based on the soft competitive learning algorithm is presented for the adaptive compression of these video signals. Based on comparisons with existing compression techniques, this investigation indicates that a compression ratio of better than 100:1 is a reasonable expectation

Keywords

adaptive signal processing; image sequences; neural nets; unsupervised learning; vector quantisation; video coding; adaptive compression algorithm; animated cartoons; animated sequences; animated video; artificial neural network system; compression ratio; soft competitive learning algorithm; speech; stationary images; unsupervised artificial neural learning algorithms; vector quantization; video signals; Animation; Artificial neural networks; Communication channels; Compression algorithms; Data compression; Image coding; Prototypes; Speech; Vector quantization; Video compression;

fLanguage

English

Publisher

ieee

Conference_Titel

WESCANEX 95. Communications, Power, and Computing. Conference Proceedings., IEEE

Conference_Location

Winnipeg, Man.

Print_ISBN

0-7803-2725-X

Type

conf

DOI

10.1109/WESCAN.1995.493969

Filename

493969