A Detail Based Method for Linear Full Reference Image Quality Prediction

TL;DR

This paper introduces a new full reference image quality assessment method that combines detail loss and spurious detail metrics, correlating well with human perception across multiple databases.

Contribution

It proposes a novel approach using gradient decomposition to separately quantify detail loss and spurious details, improving image quality prediction accuracy.

Findings

Strong correlation with empirical DMOS scores

Effective across multiple image databases

Enables DMOS scale alignment using a single noisy image

Abstract

In this paper, a novel Full Reference method is proposed for image quality assessment, using the combination of two separate metrics to measure the perceptually distinct impact of detail losses and of spurious details. To this purpose, the gradient of the impaired image is locally decomposed as a predicted version of the original gradient, plus a gradient residual. It is assumed that the detail attenuation identifies the detail loss, whereas the gradient residuals describe the spurious details. It turns out that the perceptual impact of detail losses is roughly linear with the loss of the positional Fisher information, while the perceptual impact of the spurious details is roughly proportional to a logarithmic measure of the signal to residual ratio. The affine combination of these two metrics forms a new index strongly correlated with the empirical Differential Mean Opinion Score (DMOS) for a significant class of image impairments, as verified for three independent popular databases. The method allowed alignment and merging of DMOS data coming from these different databases to a common DMOS scale by affine transformations. Unexpectedly, the DMOS scale setting is possible by the analysis of a single image affected by additive noise.