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 6

Preprint: Image Compression via Joint Statistical Model
images/spacer.gifimages/spacer.gif Reply into Digest
Previous :: Next  
Author Message
Eero Simoncelli (

PostPosted: Tue Jun 03, 1997 3:01 pm    
Subject: Preprint: Image Compression via Joint Statistical Model
Reply with quote

#2 Preprint: Image Compression via Joint Statistical Model

Dear Wavelet Digest(ees),

We've just made available a technical report preprint of a paper
describing a much-expanded version of our characterization of image
statistics in the wavelet domain, and the accompanying compression
algorithm, EPWIC. Comments/suggestions/criticisms welcome!

TITLE: Image Compression via Joint Statistical Characterization in
the Wavelet Domain

AUTHORS: Robert W. Buccigrossi and Eero P. Simoncelli


ABSTRACT: We develop a statistical characterization of natural images
in the wavelet transform domain. This characterization describes the
joint statistics between pairs of subband coefficients at adjacent
spatial locations, orientations, and scales. We observe that the raw
coefficients are nearly decorrelated, but their magnitudes are highly
correlated. A linear magnitude predictor coupled with both
multiplicative and additive uncertainties accounts for the joint
coefficient statistics of a wide variety of images including
photographic images, graphical images, and medical images. In order
to directly demonstrate the power of this model, we construct an image
coder called EPWIC (Embedded Predictive Wavelet Image Coder), in which
subband coefficients are encoded one bitplane at a time using a
non-adaptive arithmetic encoder that utilizes probabilities calculated
from the model. Bitplanes are ordered using a greedy algorithm that
considers the MSE reduction per encoded bit. The decoder uses the
statistical model to predict coefficient values based on the bits it
has received. The rate-distortion performance of the coder compares
favorably with the current best image coders in the literature.
All times are GMT + 1 Hour
Page 1 of 1

Jump to: 

disclaimer -
Powered by phpBB

This page was created in 0.024351 seconds : 18 queries executed : GZIP compression disabled