Hamid Rahim Sheikh, UT Austin, 8/26/03

An Information Theoretic Criterion for Image Quality Assessment


Measurement of quality is of fundamental importance for numerous image and video processing algorithms. The goal of Quality Assessment (QA) research is to design algorithms that can automatically assess the quality of images or videos in a perceptually consistent manner. Traditionally, image QA algorithms interpret image quality as fidelity or similarity with a `reference' or `perfect' image in some perceptual space. Such `Full-Reference' (FR) QA methods attempt to achieve consistency in quality prediction by modelling salient physiological and psychovisual features of the Human Visual System (HVS), or by arbitrary, albeit motivated, mathematical criteria. In this talk I will present my approach to the problem of image QA by proposing an information fidelity criterion that is based on Natural Scene Statistics. QA systems are invariably involved with judging the visual quality of images and videos that are meant for `human consumption'. Researchers have developed sophisticated models to capture the statistics of natural signals, that is, pictures and videos of the visual environment. Using these statistical models in an information-theoretic setting, I will present a QA algorithm that is not only parameterless, but also outperforms current methods. We also show that our approach from a diametrically opposite direction to traditional HVS based QA methods, is functionally equivalent to them under certain conditions. I will also present results that validate the performance of the algorithm with an extensive subjective study involving about 800 images.

Bio: Hamid Rahim Sheikh is a PhD candidate at the University of Texas at Austin. He is affiliated with the Laboratory for Image and Video Engineering under the supervision of Dr. Alan C. Bovik. His research interests include quality assessment, application of HVS and NSS models to image and video processing problems, video coding, and embedded implementations.