G₀ Revisited as Equally Bright Reference Boundary

Abstract

Brilliance and zero grayness (denoted as G0) and are two terms coined by Ralph Evans. Nayatani, Heckaman and Fairchild have done series of work to incorporate them into comprehensive color appearance models. In this work, those concepts were reexamined to scale lightness/brightness across the chromaticity diagram. Specifically, observers, mostly with a color science background, were asked to adjust the luminance of a color patch to appear with no grayness, or equivalently just about/cease to glow. The hypothesis was that lightness can be equalized across those chromaticities and the Helmholtz-Kohlrausch effect is automatically incorporated. This hypothesis was verified in a follow-up experiment where another group of observers completed paired comparisons of the brightness between the collected G0 results. The G0 task was also repeated under another two levels of adaption backgrounds, based on which different absolute brightness results for a given chromaticity might be derived. In addition, high correlations between the G0 results (as a perceptual boundary between appearance modes) and different physical gamut boundaries including MacAdam's optimal colors were found for possible computational proxies and ecologically meaningful implications.

The brightness of chromatic stimuli

“…Universally, the analyzed models of the H-K effect have lesser mean errors (either smaller positive or greater negative values) on the four newer data sets-our data set, the High data set, 16 the Seong data set, 25 and the Xie data set 26 -than on the three older data sets. This indicates that the strength of the H-K effect is greater in the four newer data sets than in the three older data sets.…”

“…We compared the proposed models (Equations 8 and 11) to four other models of the H-K effect 11,16,24 alongside default CIECAM16 and CIELAB on our data and five other brightness matching data sets. [9][10][11]16,25,26 These data consist of CIE XYZ values for pairs of stimuli that were matched in brightness or lightness by observers. An ideal model will produce equal lightness and brightness predictions for each pair of stimuli; the difference between predicted lightness values represents the error of the models.…”

Perceived Color Gamut in Images: From Boundary to Difference

“…Park and Murdoch ( 2020 ) studied the trade-off between chromaticity gamut area and luminance for image quality and found more chroma can equivalently compensate for the deficiency in lightness. Recent efforts have been made to address this problem by adding the compensation component in gamut mapping (Zamir et al, 2019 ) and proposing more independent color scales (Xie and Fairchild, 2021 ). Our future work includes CGV re-calculations in such a new space.…”

Virtual Colour Atlas

“…• The Virtual Colour Atlas has been recently used to aid research on the perceptions of brilliance, zero-grayness, and saturation using the unique sampling of the DIN 6164 system. 30 This has been particularly helpful since DIN 6164 atlases are long out of print and very difficult to obtain.…”