ABSTRACT This study investigates the effects of blue clothing attributes and Asian skin tones on professional impression judgments. Unlike prior research emphasizing cross‐hue comparisons, it focuses on attribute‐level variation, examining how combinations of hue, saturation, and brightness in the blue spectrum influence professional impression in business contexts. The study was conducted in two stages. First, 12 Asian skin tone samples and 20 blue clothing color samples were developed, and five affective descriptors were selected. Second, 150 participants evaluated 3D avatar stimuli combining different skin tones and blue clothing colors; the data were analyzed using descriptive statistics, ANOVA, and correlation analysis. Results indicate that clothing color attributes are the primary factors shaping professional impression, with hue exerting the strongest effect, followed by saturation–brightness (S–B) combinations. Skin tone shows a relatively small direct effect but moderates the optimal range of color configurations. High ratings are concentrated in the blue (210°) to standard blue (240°) range under medium‐to‐high saturation and brightness, whereas blue–green (190°) and blue–violet (260°) receive lower evaluations. Associations between professional impression and the affective descriptors were observed in only two of the 12 skin‐tone conditions, indicating a relative distinction between professional impression and general affective evaluations. These findings suggest that professional impression is better explained by attribute‐level configurations and their interactions with skin tone than by the category of “blue” alone, supporting attribute‐based color strategies in apparel design and professional visual communication.
ABSTRACT Camera color constancy estimates the chromaticity of the dominant scene illuminant in sensor RGB space so that illumination‐induced color casts can be corrected. Global chromatic statistics are sensitive to scene composition, while unconstrained regressors may exploit camera or dataset‐specific correlations. ChromaTokenNet addresses these limitations by processing aligned semantic and log‐chromatic patch tokens together with differentiable global log‐chromaticity histogram tokens in a shared Transformer. An illumination‐token regressor and a confidence‐pooled patch regressor produce separate log‐chromatic estimates, which are combined by a learned scalar gate. The main contribution is the way these established chromatic, attention, and reliability‐weighting principles are organized in a single model. No outer‐test result generated under that finalized protocol was used to revise the architecture or the training procedure. ChromaTokenNet achieves Mean/Best25/Median/Worst25 recovery angular errors of on ColorChecker RECommended and on INTEL–TAU. Controlled ablations, camera‐held‐out tests, perturbations, and token interventions are used to examine the contribution and limitations of the model components. Performance remains stable under intensity reduction and moderate downsampling, but declines under severe combined low‐light attenuation and additive noise, spatially varying illumination, and non‐invertible clipping.
ABSTRACT Cross‐sensor color constancy is affected by variations in sensor spectral sensitivity, often leading models trained on one sensor to degrade when applied to another. To address this challenge, we propose a physically motivated training strategy that improves cross‐sensor robustness without requiring explicit sensor characterization or multisensor training data. The proposed method incorporates the von Kries diagonal model as a dynamic training constraint. Specifically, illuminants are projected onto a diagonal manifold derived from measured ground‐truth illuminants, thereby simulating structured variations in the image formation process. This transformation preserves consistency with the radiometric model while encouraging the network to disentangle illuminant‐related factors from sensor‐dependent characteristics. As a result, a model trained on a single sensor can generalize effectively to unseen sensors. Experimental results on Cube+, INTEL‐TAU, and NUS‐8 demonstrate the effectiveness of the proposed strategy.
ABSTRACT This study characterizes the evolution of CHANEL lipstick colors from 1960 to 2020 using complete archival color charts. Because some early lipstick specimens are no longer preserved, the suitability of the charts as proxies was evaluated by comparing 146 surviving lipstick samples with their corresponding printed representations. Measurements were acquired in CIELAB using an X‐Rite VS450 noncontact spectrophotometer under 45°:0° geometry, with a 6‐mm measurement area, CIE standard illuminant D65, and the 10° standard observer. The mean CIEDE2000 color difference was 3.12 ± 2.63 (median 2.38; interquartile range 1.54–3.79; range 0.35–15.33); 64.4% and 85.6% of pairs had ΔE00 values ≤ 3 and ≤ 5, respectively. The complete set of 513 chart colors was then assigned Natural Color System (NCS) notations and analyzed by hue, blackness, and chromaticness. The results reveal a persistent predominance of yellowish‐red colors, recurrent highly chromatic reds, and comparatively few bluish‐reds. NCS blackness generally increased after 1972, with near‐black colors becoming prominent between 1995 and 2010. The number of colors expanded from 12 in 1960 to 55 in 2005 before subsequently contracting. Over time, several ranges occupied increasingly dense regions of the selected NCS representation, while differences in pearlescence, opacity, and finish provided additional lipsticks differentiation not captured by the present color‐charts analysis. These findings establish a CHANEL specific Color Time Line and demonstrate the value, as well as the limitations, of archival commercial color charts for longitudinal color research.
ABSTRACT Color matching functions (CMFs) serve as the fundamental basis for deriving tristimulus values and chromaticities, which are essential for color specification, calibration, and reproduction. With the rapid advancement of display technologies, modern commercial devices increasingly adopt displays with narrower primary spectra and larger color gamut areas approaching Rec. 2020. In this study, ten observers performed color matching using five displays selected to cover color gamuts from sRGB to those approaching Rec. 2020, with the stimuli having a field of view of 20.2°. The results showed that displays with wider color gamuts and narrower‐band primaries generally introduced greater observer metamerism and reduced the accuracy of the standard CIE CMFs in characterizing the average color matches. None of the four CIE CMFs provided accurate characterization across all displays. The CIE 2006 10° CMFs performed best for the neutral stimulus, while the CIE 1931 2° CMFs performed better for the blue and violet stimuli. For the laser projector with the narrowest primaries, all the four CIE CMFs failed to characterize color matches with average chromaticity differences greater than 0.008 Δu ′ v ′, and the observer metamerism was substantially greater than the other displays.
