RAID-Database: human Responses to Affine Image Distortions

TL;DR

A set of human responses to suprathreshold affine image transforms (rotation, translation, scaling) and Gaussian noise as convenient reference to compare with previously existing image quality databases.

Abstract

Image quality databases are used to train models for predicting subjective human perception. However, most existing databases focus on distortions commonly found in digital media and not in natural conditions. Affine transformations are particularly relevant to study, as they are among the most commonly encountered by human observers in everyday life. This Data Descriptor presents a set of human responses to suprathreshold affine image transforms (rotation, translation, scaling) and Gaussian noise as convenient reference to compare with previously existing image quality databases. The responses were measured using well established psychophysics: the Maximum Likelihood Difference Scaling method. The set contains responses to 864 distorted images. The experiments involved 105 observers and more than 20000 comparisons of quadruples of images. The quality of the dataset is ensured because (a) it reproduces the classical Piéron's law, (b) it reproduces classical absolute detection thresholds, and (c) it is consistent with conventional image quality databases but improves them according to Group-MAD experiments.

Figures (11)

• Figure 1: Example of the transformation of two reference images with the maximum level of each distortion.
• Figure 2: Example of a trial presented to the observer.
• Figure 3: MLDS responses for all distorted images. Left: all the images with its corresponding response. Right: Each of the four plots corresponds to a particular distortion, each line corresponds to the response to an image with different levels of the distortion.
• Figure 4: Raw data: Example of the database of human recordings. Only the first rows are shown. See text for details.
• Figure 5: MLDS data: Example of the database of MLDS responses computed from the raw data. Only the first rows are shown. See text for details.