ICIP 2006, Atlanta, GA
 

Slide Show

Atlanta Conv. & Vis. Bureau

 

Technical Program

Paper Detail

Paper:TA-L2.7
Session:Image Coding
Time:Tuesday, October 10, 12:00 - 12:20
Presentation: Lecture
Title: ENTROPY-BASED DISTORTION MEASURE FOR IMAGE CODING
Authors: Thomas Andre; Laboratoire I3S UNSA CNRS 
 Marc Antonini; Laboratoire I3S UNSA CNRS 
 Michel Barlaud; Laboratoire I3S UNSA CNRS 
 Robert M. Gray; Stanford University 
Abstract: Classical quality criteria for image coding are based on the mean square error, which presents many interesting characteristics but is mostly efficient at high bit-rate. We investigate here the properties of a distortion measure based on differential entropy. The proposed measure appears to lead to an interesting criterion in the non-asymptotic case and without any assumption on the probability distribution of the source. An adapted bit allocation algorithm is proposed in order to take advantage of this criterion. The first experimental results illustrate the behavior of the proposed distortion measure and exhibit interesting psycho-visual properties.