Visual consistency between display simulation and real scene observation: a case study for color preference assessment for illuminated scenes

Display-based simulation has been widely adopted across various research fields, yet the visual consistency between this technique and real scene observation remains inadequately demonstrated. This study aimed to assess the feasibility of quantifying the color rendition performance of white light sources using display simulation. We examined the visual consistency through two psychophysical experiments conducted with the assistance of spectral imaging and color management. Two real-scene experiments, a rating experiment for fruits and a pair-comparison experiment for bronzeware within viewing light booths under various light sources, were accurately simulated on displays against different colored backgrounds. The results revealed overall significant and strong correlations between the color preference assessments in real and simulated scenes. The simulation deviation we observed could be attributed to chromatic adaptation and eliminated by controlling the interface background color. Meanwhile, the evaluation methods and the observers' familiarity with the experimental objects might to some extent influence the evaluation results as well. By this study, the authors would like to highlight the crucial importance of accuracy in real scene replication for simulated reality.