Seeing the Light: Perception and Discrimination of Illumination Color.

Human perception of the color of physical surfaces is practically not affected by changes in illumination. This phenomenon is called color constancy. Based on results of previous psychophysical experiments, it has been established that there are two types of color perception: apparent color and surface color. It has also been suggested that unless there is a complete adaptation to the illuminant, color constancy can be achieved only with respect to the surface color. Computational models of color constancy boil down to problems of estimation of reflectance of the observed object based on the magnitude of the sensory response, and duality of color perception has not been adequately addressed in previous studies. This study was undertaken for the purpose of making clear the characteristics of the two types of color perception (apparent color and surface color). The experimental technique used in this study was based on the detection of changes of illuminance and reflectance for the purposes of determination of the effect of the surround stimulus on color perception, rather than on conventional color matching technique. The results of the study indicate that the surround stimulus exhibits an inhibitive influence on the color perception of the center stimulus, and the effect of the size and spatial structure of the surround stimulus is different with respect to the apparent color and the surface color. It was also demonstrated that results of the experiments can be explained by a hypothesis of a hierarchical structure of the vision system combining two different types of color perception. © 2000 Scripta Technica, Electron Comm Jpn Pt 3, 83(11): 43–55, 2000

Invariance of surface color representations across illuminant changes in the human cortex

Ratios of excitations in each cone-photoreceptor class produced by light reflected from pairs of surfaces in a scene are almost invariant under natural illuminant changes. The stability of these spatially defined ratios may explain the remarkable ability of human observers to efficiently discriminate illuminant changes from changes in surface reflectances. Spatial cone-excitation ratios are not, however, exactly invariant. This study is concerned with observers' sensitivity to these invariance violations. Simulations of Mondrian paintings with either 49 or two natural surfaces under Planckian illuminants were presented as images on a computer-controlled display in a two-interval experimental design: in one interval, the surfaces underwent an illuminant change; in the other interval, the surfaces underwent the same change but the images were then corrected so that, for each cone class, ratios of excitations were preserved exactly. Although the intervals with corrected images corresponded individually to highly improbable natural events, observers systematically misidentified them as containing the illuminant changes, the probability of error increasing as the violation of invariance in the other interval increased. For the range of illuminants and surfaces tested, sensitivity to violations of invariance was found to depend on cone class: it was greatest for long-wavelength-sensitive cones and least for short-wavelength-sensitive cones. Spatial cone-excitation ratios, or some closely related quantities, seem to be the cues preferred by observers for making inferences about surface illuminant changes.

How do human observers estimate the location, form, and color of objects? Accurate estimation is challenging because the light arriving at the eyes depends not only on object properties, but also on the spectra and spatial layout of the light sources (Nassau, 1983; Foley et al., 1990). How well the visual system separates illuminant and object properties to achieve a stable representation has traditionally been studied under the rubric of color and lightness constancy. Most previous work used very simple stimuli, typically a few diffusely illuminated surfaces arranged perpendicular to the line of sight. Over the past several years, however, there has been an evident increase of interest in expanding the conceptualization of this area to incorporate effects that emerge only for complex, typically three-dimensional, scenes. The current issue features papers that represent various manifestations of this interest. One line of research investigates how the three-dimensional layout of a scene affects the perception of lightness and color. Although the current work has long-standing antecedents (e.g. Mach, 1886/1959; Hochberg and Beck, 1954; Gilchrist, 1980), methodological advances in i) experimentation with real illuminated objects (e.g. Brainard et. al, 1987; Rutherford and Brainard, 2002; Ripamonti et al., 2004; Robilotto and Zaidi, 2004), ii) the use of sophisticated graphics simulations (e.g. Yang and Maloney, 1999; Fleming, Dror, & Adelson, 2003; Boyaci, Maloney, & Hersh, 2003; Delahunt and Brainard, 2004), iii) the design of hybrid systems that combine real objects with image-based graphics and video projection (Ling and Hurlbert, 2004), and iv) psychophysical procedures (Maloney and Yang, 2003; Obein, Knoblauch, & Vienot, 2004) have opened the door for systematic exploration of a wider range of phenomena. Recent papers include work on how well vision compensates for changes in surface orientation (Boyaci et al., 2003; Ripamonti et al., 2004), how effectively it discounts inter-reflections among nearby surfaces (Bloj, Kersten, & Hurlbert, 1999; Doerschner, Boyaci, & Maloney, 2004; Delahunt and Brainard, 2004), and how the visual system effectively estimates the spectral properties and spatial layout of the illuminant in three-dimensional scenes (Kraft & Brainard, 1999; Yang & Maloney, 1999; Boyaci, Maloney, & Hersh, 2003; Bloj et al., 2004; Boyaci, Doerschner, & Maloney, 2004; Khang and Zaidi, 2004). The second thread that leads to papers in the current issue is a focus on the functional utility of color and lightness perception -- the idea that these percepts inform us about the properties of objects rather than those of light spectra. This focus resulted in a renaissance of research in color constancy over the past two decades, with particular progress being made in the development of computational models that explore how, in principle, object surface properties can be estimated from image data. As with the experimental lines, early work focused on simple scene geometries (for reviews see Hurlbert, 1998; Maloney, 1999) but consideration has recently expanded to three-dimensional configurations (Adelson and Pentland, 1996; Yang and Maloney, 1999; Bell and Freeman, 2001; Dror, Willsky, & Adelson, 2004) Of particular interest has been the elaboration of purely computational formulations into parametric models of human performance (e.g. Brainard Brunt, & Speigle, 1997; Brainard Kraft, & Longere, 2003; Boyaci et al., 2003; Doerschner et al., 2004; Boyaci et al., 2004; Bloj et al., 2004), tests of how well the visual system exploits image information identified in computational studies (Yang and Maloney, 2001; Delahunt and Brainard, 2004; Smithson and Zaidi, 2004), investigations of how well the visual system recovers perceptual correlates of material properties other than diffuse surface reflectance, such as gloss and translucency (Lu, Koenderink, & Kappers, 2000; Fleming et al., 2003; Pont & Koenderink, 2003; Obein et al., 2004), as well as how geometric aspects of surface reflectance interact with the perception of shape (Fleming et al., 2003).

The disagreement between people who named #theDress (the Internet phenomenon of 2015) “blue and black” versus “white and gold” is thought to be caused by individual differences in color constancy. It is hypothesized that observers infer different incident illuminations, relying on illumination “priors” to overcome the ambiguity of the image. Different experiences may drive the formation of different illumination priors, and these may be indicated by differences in chronotype. We assess this hypothesis, asking whether matches to perceived illumination in the image and/or perceived dress colors relate to scores on the morningness-eveningness questionnaire (a measure of chronotype). We find moderate correlations between chronotype and illumination matches (morning types giving bluer illumination matches than evening types) and chronotype and dress body matches, but these are significant only at the 10% level. Further, although inferred illumination chromaticity in the image explains variation in the color matches to the dress (confirming the color constancy hypothesis), color constancy thresholds obtained using an established illumination discrimination task are not related to dress color perception. We also find achromatic settings depend on luminance, suggesting that subjective white point differences may explain the variation in dress color perception only if settings are made at individually tailored luminance levels. The results of such achromatic settings are inconsistent with their assumed correspondence to perceived illumination. Finally, our results suggest that perception and naming are disconnected, with observers reporting different color names for the dress photograph and their isolated color matches, the latter best capturing the variation in the matches.

Simple constraints on the sets of possible surface reflectances and illuminants are exploited in a new color constancy algorithm that builds upon Forsyth's (1990) theory of color constancy. The goal defined for a color constancy algorithm is to discount variations in the color and intensity of the incident illumination and thereby extract illumination-independent descriptors of surface colors from images. Forsyth's method is based on two constraints: first, the surface colors under a canonical illuminant all fall within an established maximal convex gamut of possible colors and second that a matrix accurately maps colors between illuminants. These constraints taken together turn out to be very effective in solving for color constancy; however, other strong assumptions about the scenes are required for the method to work-the illumination must be uniform, the surfaces must be planar, and there can be no specularities. We show that these restrictions are necessary only because Forsyth sets out to recover the intensity of descriptors. At the outset we abandon 3-dimensional descriptor recovery in favor of recovering only orientation (i.e. 2 dimensions). Intensity information is factored out of the problem by mapping 3-dimensional (r, g, b) camera responses onto 2-dimensional chromaticities; specifically (r/b, g/b). We show that this diagonal chromaticity has two important properties: first, gamut convexity is preserved and second illumination change is still described by a matrix. It follows that Forsyth's algorithm can be directly applied to the recover chromaticity descriptors and from these the 3D descriptor orientations can be derived. The basic algorithm is then extended to include a maximal gamut constraint on the set of illuminants that is analogous to the gamut constraint on surface colors. The chromaticity space facilitates the expression of the illumination constraint in the algorithm. Tests on real images show that the algorithm provides good color constancy. >

Asymmetries in blue–yellow color perception and in the color of ‘the dress’

Perception of colour depends on the spectral composition of light that reaches retina and depends as well on various mechanisms of visual system that processes information flow. The few important mechanisms can be distinguished in colour perception: colour adaptation, colour constancy and colour contrast. If the visual field has only one coloured object, then colour perception will be determined by spectral composition of the light and colour adaptation. Whereas mechanisms of colour constancy and colour contrast switches on when in the visual field there are at least two colour objects. Von Kries (1905) have attributed colour constancy phenomenon to theory of receptoral adaptation. But this is theory is applicable to local processes happening in relatively small size of visual field. Craven and Foster (1992) shows that receptor excitation ratios remains constant during change of illumination. But again remains unexplained colour change during adaptation. So, the hypothesis is raised that two processes attribute colour perception: local colour contrast calculation and global adaptation to the background. Therefore the experiments have been carried out to establish the colour contrast and background adaptation impact on colour perception.Four subjects with normal colour vision participated in the experiments. 40 Munsell samples (value 7 and chroma 4) illuminated with one reference illuminant (standard C) and 6 test illuminants (standard A, standard S, cardinal red, cardinal green, cardinal yellow and cardinal blue) were simulated on computer monitor. The dark-adapted subjects have been shown colour samples on neutral background illuminated with one of test illuminants. The task was to match the colour stimuli appearance under reference illuminant. The sequential asymmetric matching procedure was carried out. First subject adapts to neutral background under illuminan C then test stimulus and test background appears for limited time followed by readaptation to neutral background. Then subject is asked to adjust the colour mach under illuminant C. Two different experiment paradigms have been used. In the first one the stimulus had 2° on neutral background covering 20° of visual field. In the second paradigm the stimulus size was the same but the background was covering the all visual field. The adaptation and presentation timings were 1, 5, 30 or 60 seconds.Results under the first experiment paradigm show partial colour constancy for all subjects and for all test illuminants under various adaptation times. Despite quite a long adaptation time of 30 seconds subjects was unable to achieve full colour constancy. Colour constancy improves under second experiment paradigm. 60-second adaptation time is enough to achieve full colour constancy. The less adaptation time (1 or 30 second) gives partial colour constancy.The subject sees two colours (stimulus and background) during the experiment. The colour perception of stimulus and background changes during adaptation time. In the first 5 seconds of adaptation 40–60% of full adaptation level is reached (Fairchild and Reniff, 1995; Werner et al., 2000). Wesner and Shevell (1992, 1994) shown that signals from all visual field have influence on colour perception of object. The first part of our experiment shows that neutral background of 20° gives partial adaptation to background colour. Therefore, increase background illumination to full visual field gives full adaptation to background. So, subject perceives different coloured background under various illuminants as the same neutral (grey) background.We can state that two different systems have been distinguished in the process of colour perception. One system evaluates the colour difference between the stimulus and the background. Second system evaluates the colour of the background. Perceived colour difference during adaptation does not change much, but perceived colour of the background changes a lot and drifts towards neutral colour. The hypothesis of two level colour perception is supported by experimental data.

In our previous work [J. Opt. Soc. Am. A42, B81 (2025)JOAOD60740-323210.1364/JOSAA.545278], we demonstrated that the spatial local contrast influences color constancy performance by combining the effects of illuminant colors and background surface colors in two-dimensional (2D) stimuli. In this study, we investigated the impact of perceived scene dimensionality, achieved by using the perspective space technique in which a lighting box filled with a chromatic illuminant was simulated (abbreviated as 2D+ scene) on color constancy. White daylight (D65) and red, green, blue, and yellow chromatic illuminants illuminated the reference and test stimuli in a haploscopic view, respectively. Each stimulus consisted of a central 1.2° test patch surrounded by a 4.2° background of Mondrian-like overlapping chromatic patches and an outer 0.4° gray fringe. On a neutral background, the colors of the patches were in the L ∗ a ∗ b ∗ hue circle around gray, and on chromatic backgrounds dominated by red, green, blue, and yellow, these patch colors were systematically shifted to one color direction in maintaining the hue circle. In the first two conditions, both the reference and test stimuli had the same background (neutral and complementary to the illuminant color). In contrast, in the last two conditions, in which both the illuminant color and the background dominant colors were systematically controlled, the reference and test stimuli had different backgrounds, and the color appearance of the backgrounds, except at the fringe, looked similar (the color of the illuminant and neutral). The results showed that in the 2D scene, the color constancy remarkably declined with systematic surface color changes under the last two conditions. On the contrary, in the 2D+ scene, this decline was strongly moderated, although the decline was still statistically significant, except for the green illuminant. These results suggest that if the illuminant color is perceived as a 3D light source in the 2D+ scene, the effectiveness of the illumination estimation is recovered, and the local contrast mechanism still partially contributes to color constancy. However, the simultaneous and systematic color shifts in illuminants and background surfaces still reduced the color constancy performance, despite the lighting box and fringe providing clear and consistent cues to estimate the illuminant colors.

Color constancy is the ability to recognize the color of an object (or more generally of a surface) under different illuminations. Without color constancy, surface color as a perceptual attribute would not be meaningful in the visual environment, where illumination changes all the time. Nevertheless, it is not obvious how color constancy is possible in the light of metamer mismatching. Surfaces that produce exactly the same sensory color signal under one illumination (metamerism) may produce utterly different sensory signals under another illumination (metamer mismatching). Here we show that this phenomenon explains to a large extent the variation of color constancy across different colors. For this purpose, color constancy was measured for different colors in an asymmetric matching task with photorealistic images. Color constancy performance was strongly correlated to the size of metamer mismatch volumes, which describe the uncertainty of the sensory signal due to metamer mismatching for a given color. The higher the uncertainty of the sensory signal, the lower the observers' color constancy. At the same time, sensory singularities, color categories, and cone ratios did not affect color constancy. The present findings do not only provide considerable insight into the determinants of color constancy, they also show that metamer mismatch volumes must be taken into account when investigating color as a perceptual property of objects and surfaces.

Colour vision: Putting it in context

The stable perception of surface colors in variable viewing conditions is important for information about surface color to help in object identification. This is a difficult computational problem, yet people are good at judging the colors of objects and materials under changing lighting conditions. Mechanisms at different levels of visual information processing work together to achieve this color constancy. Much is known about the sensory mechanisms of color constancy, whereas the role of higher-level mechanisms is far less understood. This chapter focuses on the influence of previous experience, specifically memory colors, on color appearance and color constancy. Previously used methodologies are discussed, in particular whether they are capable of discerning between semantic and visual effects in memory color. The chapter presents recent work showing that prior experience can affect the color appearance of natural objects. This could provide an additional, high-level cue for achieving color constancy.

The color of an object, when part of a complex scene, is determined not only by its spectral reflectance but also by the colors of all other objects in the scene (von Helmholtz, 1886; Ives, 1912; Land, 1959). By taking global color information into account, the visual system is able to maintain constancy of the color appearance of the object, despite large variations in the light incident on the retina arising from changes in the spectral content of the illuminating light (Hurlbert, 1998; Maloney, 1999). The neural basis of this color constancy is, however, poorly understood. Although there seems to be a prominent role for retinal, cone-specific adaptation mechanisms (von Kries, 1902; Pöppel, 1986; Foster and Nascimento, 1994), the contribution of cortical mechanisms to color constancy is still unclear (Land et al., 1983; D'Zmura and Lennie, 1986). We examined the color perception of 27 patients with defined unilateral lesions mainly located in the parieto-temporo-occipital and fronto-parieto-temporal cortex. With a battery of clinical and specially designed color vision tests we tried to detect and differentiate between possible deficits in central color processing. Our results show that color constancy can be selectively impaired after circumscribed unilateral lesions in parieto-temporal cortex of the left or right hemisphere. Five of 27 patients exhibited significant deficits in a color constancy task, but all of the 5 performed well in color discrimination or higher-level visual tasks, such as the association of colors with familiar objects. These results indicate that the computations underlying color constancy are mediated by specialized cortical circuitry, which is independent of the neural substrate for color discrimination and for assigning colors to objects.

Color Constancy for an Unseen Surface

We simultaneously perceive at each visible surface point (at least) a surface color (with several dimensions), an illuminant color, surface orientation, and surface specularity. Perceived surface and lighting variables are a multidimensional function of the past several minutes of chromatic exposure and the current retinal images. Models with outputs that are single-valued functions of image spatial and temporal derivatives (e.g., edge ratio models of Wallach, Land, Horn, Arend) describe sensory processes providing important relational information. However, they are easily shown to be incomplete as models of surface color constancy. Traditional color appearance models are similarly incomplete. This multidimensionality of surface perception was known to early theorists and has been further elaborated by recent human perception and machine vision theorists. The human vision data needed for refinement of computational color constancy models are currently in short supply for both theoretical and methodological reasons. We have recently measured human color constancy using both spatially simple and complex patterns. Our procedure allowed separate measurement of local sensory color (hue, saturation, and brightness) and apparent surface colors (surface chromatic color and lightness). Even in the simplified scenes of our experiments there is a complicated relationship between the sensory color at an image point and the apparent color of the surface perceived on the corresponding sight line.