Back to articles
Articles
Volume: 13 | Article ID: art00002
Image
Image Preference Scaling for HDR Image Rendering
Jiangtao Kuang
Garrett M. Johnson
Mark D. Fairchild
  DOI :  10.2352/CIC.2005.13.1.art00002  Published Online :  January 2005
Abstract

Two psychophysical experiments were performed to evaluate image preference of 6 high-dynamic-range (HDR) image rendering algorithms. The experiments were split into a paired comparison experiment examining overall preference, and a rating scale experiment judging individual preference for 6 image attributes: highlight details, shadow details, overall contrast, sharpness, colorfulness and artifacts. The paired comparison experiment was analyzed using Thurstone's law to generate interval scales. In addition, dual scaling analysis indicates a single perceptual dimension accounting for the variance of overall preference. The overall preference shows high correlations with shadow details, overall contrast, sharpness and colorfulness, which represent the most important factors in observers' preference judgment. Stepwise regression of various image attributes to the overall preference results showed that for many images the preference scales of a single attribute can predict the overall image preference.

Journal Title : Color and Imaging Conference
Publisher Name : Society of Imaging Science and Technology
Publisher Location : 7003 Kilworth Lane, Springfield, VA 22151, USA
Subject Areas :
Views 14
Downloads 0
 DOI 10.2352/CIC.2005.13.1.art00002
 articleview.views 14
 articleview.downloads 0
  Cite this article 

Jiangtao Kuang, Garrett M. Johnson, Mark D. Fairchild, "Image Preference Scaling for HDR Image Rendering"  in Proc. IS&T 13th Color and Imaging Conf.,  2005,  pp 8 - 13,  https://doi.org/10.2352/CIC.2005.13.1.art00002

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2005
72010350
Color and Imaging Conference
color imaging conf
2166-9635
Society of Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151, USA
2166-9635(20050101)2005:1L.8;1-
cic_v2005n1/splitsection2.xml
/ist/cic/2005/00002005/00000001/art00002
Articles
Image Preference Scaling for HDR Image Rendering
KuangJiangtao
JohnsonGarrett M.
FairchildMark D.
01012005
2005
1
8
13
2005
Two psychophysical experiments were performed to evaluate image preference of 6 high-dynamic-range (HDR) image rendering algorithms. The experiments were split into a paired comparison experiment examining overall preference, and a rating scale experiment judging individual preference for 6 image attributes: highlight details, shadow details, overall contrast, sharpness, colorfulness and artifacts. The paired comparison experiment was analyzed using Thurstone's law to generate interval scales. In addition, dual scaling analysis indicates a single perceptual dimension accounting for the variance of overall preference. The overall preference shows high correlations with shadow details, overall contrast, sharpness and colorfulness, which represent the most important factors in observers' preference judgment. Stepwise regression of various image attributes to the overall preference results showed that for many images the preference scales of a single attribute can predict the overall image preference.