Eric C. Larson
Associate Professor
Computer Science
Bobby B. Lyle School of Engineering
Southern Methodist University

twitter: @ec_larson

CS Office: 451 Caruth Hall
Lyle School of Engineering
Caruth Hall
3145 Dyer Street, Suite 445
Dallas, TX 75205

SMU UbiComp Lab:
Johnson Square 189

Categorical Subjective Image Quality (CSIQ)
I made this dataset while working at the The Image Coding and Analysis Lab at Oklahoma State University. The CSIQ database consists of 30 original images, each is distorted using six different types of distortions at four to five different levels of distortion. CSIQ images are subjectively rated base on a linear displacement of the images across four calibrated LCD monitors placed side by side with equal viewing distance to the observer. The database contains 5000 subjective ratings from 35 different observers, and ratings are reported in the form of DMOS.

The files can be downloaded in multiple parts:
When using the Database, please cite the following:
  • E. C. Larson and D. M. Chandler, "Most apparent distortion: full-reference image quality assessment and the role of strategy," Journal of Electronic Imaging, 19 (1), March 2010. Full Text PDF
A recent implementation of MAD was created in PyTorch using various GPU speed-ups that are now available from such deep learning modules. That code is available as described below:
  • Source Code.
  • Ding, K., Ma, K., Wang, S. et al. Comparison of Full-Reference Image Quality Models for Optimization of Image Processing Systems. Int J Comput Vis 129, 1258–1281 (2021). DOI

Copyright (c) 2013 Eric Larson, All rights reserved. Design by Many design elements on this site are courtesy of Jon Froehlich.