SSIM

SSIM stands for Structural SIMilarity index (it was originally known as University Quality Index, UQI). SSIM is an algorithm proposed by LIVE. It is a quality metric for images and it was developed from intuitive ideas regarding how distortions in image structure might be measured.

There are several papers that describe SSIM (and the improved versions of this metric). However, I like one of the earlier papers because it is a very straightforward description of this technique:

A universal image quality index
Zhou Wang; Bovik, A.C.;
Signal Processing Letters, IEEE
Volume 9, Issue 3, March 2002 Page(s):81 - 84

I think this is a very good starting point for anyone interested in image and video quality assessment (QA). I also found a Keynote Address paper by Bovik's group. It provides a good overview of QA.

Although these papers might provide a good introduction to QA, there is still the challenge of creating efficient quality metrics that allow us to compare HDR versus LDR content. The following paper deals with this problem:

Aydin, T. O., Mantiuk, R., Myszkowski, K., and Seidel, H. 2008. Dynamic range independent image quality assessment. In ACM SIGGRAPH 2008 Papers (Los Angeles, California, August 11 - 15, 2008). SIGGRAPH '08. ACM, New York, NY, 1-10.