Measuring the Effect of High-Level Visual Masking in Subjective Image Quality Assessment with Priming

Abstract

Despite recent advances in subjective image quality research, this topic still calls for many fundamental questions. Though we understand several early vision mechanisms fairly well, little is known about late vision and how the two interact with each other. Here, we look at one particular limitation of our visual system that stems from failures of that interaction: high-level visual masking, best illustrated by the change blindness paradigm. We carried out a user study designed specifically to measure the influence of high-level masking by priming. Results suggest a significant influence of high-level masking in image fidelity assessment at the 95% confidence level for half of the participants, with an average magnitude over three times that of intra-observer variability.

High-Level Visual Masking of Image Compression Artefacts

“…the early visual system. However, in a recent series of publications [1,2], we demonstrated a significant effect of higher-level mechanisms of visual working memory in masking effects. We have proposed that visual masking can be of two distinct kinds: low-level (masking occurs in early vision and the masked target cannot be seen even though one knows where it is) and high-level (masking occurs beyond early vision and the masked target cannot be seen until the observer knows where it is).…”

Subjective Image Fidelity Assessment: Effect of the Spatial Distance Between Stimuli

“…Generally, the main challenge is predicting whether artefacts due to the reproduction process (coding, gamut mapping, noise, blur...) are perceptible to the extent that they influence our overall judgment. One key paradigm that has been exploited in this context is masking, which can be of two kinds [9]: low-level masking prevents the perception of differences between stimuli even though we know where they Fig. 1.…”