From mixtures of mixtures to adaptive transform coding

Cynthia Archer, Todd K. Leen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

We establish a principled framework for adaptive transform coding. Transform coders are often constructed by concatenating an ad hoc choice of transform with suboptimal bit allocation and quantizer design. Instead, we start from a probabilistic latent variable model in the form of a mixture of constrained Gaussian mixtures. From this model we derive a transform coding algorithm, which is a constrained version of the generalized Lloyd algorithm for vector quat design. A byproduct o our derivatn is the introduction of a new transform basis, which unlike other transforms (PCA DCT, etc.) is explicitly optimized for coding. Image compression experiments show adaptive transform coders designed with our algorithm improve compressed image signal-to-noise ratio up to 3 dB compared to global transform codi and 0.5 to 2 dB compared to other adaptive trm coderes.

Original languageEnglish (US)
Title of host publicationAdvances in Neural Information Processing Systems
PublisherNeural information processing systems foundation
ISBN (Print)0262122413, 9780262122412
StatePublished - 2001
Event14th Annual Neural Information Processing Systems Conference, NIPS 2000 - Denver, CO, United States
Duration: Nov 27 2000Dec 2 2000

Other

Other14th Annual Neural Information Processing Systems Conference, NIPS 2000
CountryUnited States
CityDenver, CO
Period11/27/0012/2/00

Fingerprint

Image compression
Byproducts
Signal to noise ratio
Experiments

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

Archer, C., & Leen, T. K. (2001). From mixtures of mixtures to adaptive transform coding. In Advances in Neural Information Processing Systems Neural information processing systems foundation.

From mixtures of mixtures to adaptive transform coding. / Archer, Cynthia; Leen, Todd K.

Advances in Neural Information Processing Systems. Neural information processing systems foundation, 2001.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Archer, C & Leen, TK 2001, From mixtures of mixtures to adaptive transform coding. in Advances in Neural Information Processing Systems. Neural information processing systems foundation, 14th Annual Neural Information Processing Systems Conference, NIPS 2000, Denver, CO, United States, 11/27/00.
Archer C, Leen TK. From mixtures of mixtures to adaptive transform coding. In Advances in Neural Information Processing Systems. Neural information processing systems foundation. 2001
Archer, Cynthia ; Leen, Todd K. / From mixtures of mixtures to adaptive transform coding. Advances in Neural Information Processing Systems. Neural information processing systems foundation, 2001.
@inproceedings{103d244c27f94cdbbf44c62c4253b7e1,
title = "From mixtures of mixtures to adaptive transform coding",
abstract = "We establish a principled framework for adaptive transform coding. Transform coders are often constructed by concatenating an ad hoc choice of transform with suboptimal bit allocation and quantizer design. Instead, we start from a probabilistic latent variable model in the form of a mixture of constrained Gaussian mixtures. From this model we derive a transform coding algorithm, which is a constrained version of the generalized Lloyd algorithm for vector quat design. A byproduct o our derivatn is the introduction of a new transform basis, which unlike other transforms (PCA DCT, etc.) is explicitly optimized for coding. Image compression experiments show adaptive transform coders designed with our algorithm improve compressed image signal-to-noise ratio up to 3 dB compared to global transform codi and 0.5 to 2 dB compared to other adaptive trm coderes.",
author = "Cynthia Archer and Leen, {Todd K.}",
year = "2001",
language = "English (US)",
isbn = "0262122413",
booktitle = "Advances in Neural Information Processing Systems",
publisher = "Neural information processing systems foundation",

}

TY - GEN

T1 - From mixtures of mixtures to adaptive transform coding

AU - Archer, Cynthia

AU - Leen, Todd K.

PY - 2001

Y1 - 2001

N2 - We establish a principled framework for adaptive transform coding. Transform coders are often constructed by concatenating an ad hoc choice of transform with suboptimal bit allocation and quantizer design. Instead, we start from a probabilistic latent variable model in the form of a mixture of constrained Gaussian mixtures. From this model we derive a transform coding algorithm, which is a constrained version of the generalized Lloyd algorithm for vector quat design. A byproduct o our derivatn is the introduction of a new transform basis, which unlike other transforms (PCA DCT, etc.) is explicitly optimized for coding. Image compression experiments show adaptive transform coders designed with our algorithm improve compressed image signal-to-noise ratio up to 3 dB compared to global transform codi and 0.5 to 2 dB compared to other adaptive trm coderes.

AB - We establish a principled framework for adaptive transform coding. Transform coders are often constructed by concatenating an ad hoc choice of transform with suboptimal bit allocation and quantizer design. Instead, we start from a probabilistic latent variable model in the form of a mixture of constrained Gaussian mixtures. From this model we derive a transform coding algorithm, which is a constrained version of the generalized Lloyd algorithm for vector quat design. A byproduct o our derivatn is the introduction of a new transform basis, which unlike other transforms (PCA DCT, etc.) is explicitly optimized for coding. Image compression experiments show adaptive transform coders designed with our algorithm improve compressed image signal-to-noise ratio up to 3 dB compared to global transform codi and 0.5 to 2 dB compared to other adaptive trm coderes.

UR - http://www.scopus.com/inward/record.url?scp=0004921502&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0004921502&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0262122413

SN - 9780262122412

BT - Advances in Neural Information Processing Systems

PB - Neural information processing systems foundation

ER -