Crystal or jelly? Effect of color on the perception of translucent materials with photographs of real-world objects

Abstract

Translucent materials are ubiquitous in nature (e.g. teeth, food, and wax), but our understanding of translucency perception is limited. Previous work in translucency perception has mainly used monochromatic rendered images as stimuli, which are restricted by their diversity and realism. Here, we measure translucency perception with photographs of real-world objects. Specifically, we use three behavior tasks: binary classification of “translucent” versus “opaque,” semantic attribute rating of perceptual qualities (see-throughness, glossiness, softness, glow, and density), and material categorization. Two different groups of observers finish the three tasks with color or grayscale images. We find that observers’ agreements depend on the physical material properties of the objects such that translucent materials generate more interobserver disagreements. Further, there are more disagreements among observers in the grayscale condition in comparison to that in the color condition. We also discover that converting images to grayscale substantially affects the distributions of attribute ratings for some images. Furthermore, ratings of see-throughness, glossiness, and glow could predict individual observers’ binary classification of images in both grayscale and color conditions. Last, converting images to grayscale alters the perceived material categories for some images such that observers tend to misjudge images of food as non-food and vice versa. Our result demonstrates that color is informative about material property estimation and recognition. Meanwhile, our analysis shows that mid-level semantic estimation of material attributes might be closely related to high-level material recognition. We also discuss individual differences in our results and highlight the importance of such consideration in material perception.

Unsupervised learning reveals interpretable latent representations for translucency perception

“…Meanwhile, the material attributes perceived by the observers are highly positively correlated with one another for both real photographs and generated images of soaps (Figure 3(c)). The correlation among the attributes are in agreement with our previous empirical findings 12 . Figure 3(d) shows examples of real and generated images judged to have various degrees of translucency similar to that of real photographs.…”

“…We also discover the intricate role of color in translucent appearance. Most of the previous works explored the effect of color on material perception and recognition by manipulating the color/luminance distribution of material images 12,21,63,70,111,112 . For example, converting color translucent images to grayscale ones decreases perceived translucency 12,21,63,71 .…”

“…Our ICA analysis shows that oriented chromatic kernels can detect complex patterns along the translucent edges (see Figure 7(d) row 2, column 4). In addition, such chromatic kernels also capture the effect of "glow", an important feature characterizing the spatial distribution of color of translucent materials 12,21 . Furthermore, our results indicate that the presence of caustic patterns can be an important cue for translucency perception 110 .…”

“…In Experiment 2, we evaluated whether the generated images of soaps could convey perceptually persuasive material 4/29 qualities. Specifically, the same twenty observers from Experiment 1 rated three translucency-related attributes on a seven-point scale (1 means low, 7 means high): translucency, see-throughness, and glow (Figure 3(a)), which were found in a previous study to be descriptive in semantic judgments of translucent objects 12 . Material attribute ratings were normalized to the range 0 to 1 for each observer.…”

Visual discrimination of optical material properties: A large-scale study

“…Such estimation errors have provided useful information as to what kind of image cues humans use to estimate gloss. A significant number of psychophysical studies have been carried out not only on gloss but also on other optical material properties (e.g., translucency, transparency, wetness) ( Fleming & Bulthoff, 2005 ; Motoyoshi, 2010 ; Xiao et al., 2014 ; Sawayama, Adelson, & Nishida, 2017 a, Liao, Sawayama, & Xiao, 2022 ) and mechanical material properties (e.g., viscosity, elasticity) ( Kawabe, Maruya, Fleming, & Nishida, 2015 ; Paulun et al., 2017 ; van Assen, Barla & Fleming, 2018 ; van Assen, Nishida, & Fleming, 2020 ). Neurophysiological and neuroimaging studies have revealed various neural mechanisms underlying material perception ( Kentridge, Thomson, & Heywood, 2012 ; Nishio, Goda, & Komatsu, 2012 ; Nishio, Shimokawa, Goda, & Komatsu, 2014 ; Miyakawa et al., 2017 ).…”