The Wavelet Digest Homepage
Return to the homepage
Search the complete Wavelet Digest database
Help about the Wavelet Digest mailing list
About the Wavelet Digest
The Digest The Community
 Latest Issue  Back Issues  Events  Gallery
The Wavelet Digest
   -> Volume 6, Issue 4

Preprint: Image coding based on Mixture Modeling
images/spacer.gifimages/spacer.gif Reply into Digest
Previous :: Next  
Author Message
Kannan Ramchandran (

PostPosted: Sat Mar 08, 1997 1:37 am    
Subject: Preprint: Image coding based on Mixture Modeling
Reply with quote

#5 Preprint: Image coding based on Mixture Modeling

Image Coding based on Mixture Modeling of Wavelet Coefficients
and a Fast Estimation-Quantization Framework

Scott M. Lopresto and Kannan Ramchandran,
University of Illinois at Urbana-Champaign


Michael T. Orchard,
Princeton University.

We introduce a new image compression paradigm that combines
compression efficiency with speed, and is based on an independent
infinite mixture model which accurately captures the space-frequency
characterization of the wavelet image representation. Specifically,
we model image wavelet coefficients as being drawn from an independent
Generalized Gaussian distribution field, of fixed unknown shape for
each subband, having zero mean and unknown slowly spatially-varying
variances. Based on this model, we develop a powerful ``on the fly"
Estimation-Quantization (EQ) framework that consists of: (i) first
finding the Maximum-Likelihood estimate of the individual
spatially-varying coefficient field variances based on causal and
quantized spatial neighborhood contexts; and (ii) then applying an
off-line Rate-Distortion (R-D) optimized quantization/entropy coding
strategy, implemented as a fast lookup table, that is optimally
matched to the derived variance estimates. A distinctive feature of
our pardigm is the dynamic switching between forward and backward
adaptation modes based on the perceived reliability of causal
prediction contexts. The performance of our coder is extremely
competitive with the best published results in the wavelet coding
literature across diverse classes of images and target bitrates of
interest, in both compression efficiency and processing speed. For
example, our coder exceeds the objective performance of the R-D
optimized zerotree wavelet coder (based on
Space-Frequency-Quantization) at all bit rates for all tested images
at a fraction of its complexity. At low to medium bit rates (at or
below about 0.25 bpp) our preliminary results appear to outperform
almost all reported results in the literature, to the best of our
All times are GMT + 1 Hour
Page 1 of 1

Jump to: 

disclaimer -
Powered by phpBB

This page was created in 0.025392 seconds : 18 queries executed : GZIP compression disabled