An investigation of color difference for binocular rivalry and a preliminary rivalry metric, ΔE*bino

Yuta Asano,Minqi Wang
DOI: https://doi.org/10.1002/col.22900
2023-09-09
Abstract:Performance of four dE metrics (ΔE*ab, ΔE*94, ΔE00, ΔE*bino) as a binocular rivalry predictor, expressed in receiver operator characteristic (ROC) curves and area under the ROC curve (AUC) using derivation dataset. Head‐mounted displays (HMDs) for virtual and augmented reality applications could have severe nonuniformities due to complex optical components and the choice of light source technologies. In addition, module variations between the two eyes can result in different colors being seen with the left and right eyes. Binocular rivalry is expected for large color differences between the two eyes, which can severely affect product usability. Thus, it is important to characterize and predict the binocular rivalry in HMDs caused by interocular color differences. As a step toward this end, we propose the first binocular color difference metric, ΔE*bino, to predict the just noticeable binocular rivalry for a pair of solid colors. A psychophysical experiment was conducted using Meta Quest 2, in which we presented different colors to the left and right eye. The collected data were analyzed to derive the metric using ROC curves. We show that the ΔE*bino metric showed significant improvements over conventional color difference formulae. Both the ΔE*bino formula and the anonymized experiment results are available for download.
imaging science & photographic technology,chemistry, applied,optics
What problem does this paper attempt to address?