Existing color constancy methods based on deep learning primarily rely on the RGB domain and often struggle with accurate illuminant estimation in scenes with minimal spatial information, such as monochromatic environments, leading to suboptimal performance. To address this issue, this paper introduces an approach that utilizes multispectral (MS) images estimated by a pretrained RGB-to-MS model, enabling more accurate illuminant estimation. Additionally, we propose a graph-based spectral attention mechanism designed to effectively extract spectral features within the MS domain, further enhancing the robustness and accuracy of color constancy. This approach demonstrates outstanding effectiveness on our custom dataset, significantly outperforming existing methods. Additionally, when evaluated in the widely recognized NUS-8 and Cube+ datasets, the proposed method shows a substantial relative improvement of 11.5% in NUS-8 and 9.9% in Cube+ compared to previous state-of-the-art methods. Our codes and dataset will be updated at : https://github.com/sy98baek/pgsac.git

This paper presents an empirical investigation into illuminant estimation using multi-spectral images. Our study emphasizes two key contributions: (1) the utilization of the estimated multi-spectral images and (2) the incorporation of a hierarchical structure. Firstly, exploiting multi-spectral images proves to have a positive influence on illuminant estimation, particularly in scenarios characterized by monochromatic images where conventional color constancy methods face challenges. Our experimental results vividly illustrate the effectiveness of leveraging spectral information in enhancing illuminant estimation. Secondly, the adoption of a hierarchical structure stems from the need for spatial invariance in the task of estimating a global illuminant. To further enhance the performance of the hierarchical structure, we employ a contrastive loss applied to different scaled outputs. This approach demonstrates remarkable effectiveness on our custom dataset, showcasing superior performance compared to the existing methods. In addition, we extend the evaluation to the widely recognized NUS-8 dataset, where the proposed method showcases a notable 26.7% relative improvement over the previous state-of-the-art methods.

The cognitive strength of crosstalk in stereoscopic 3-D displays is investigated, and new quantitative analysis methods based on color difference and grayscale levels are developed. Unlike results using existing metrics, results by the new methods agree well with the perceived crosstalk strength in achromatic images with various levels of grayscale. The crosstalk in color images, which has not been studied before, exhibits interesting results in that the crosstalk metric based on the lightness difference expresses the best fit with the perceptual crosstalk when the intended image is black and the chroma value of the counterpart image is large, but the metric using the color difference works better when the intended image is not black. The new metrics reveal that the difference between active and passive 3-D displays is not as large as suggested by conventional crosstalk metrics, and the crosstalk in color images cannot be simply estimated by averaging the crosstalk of red, green, and blue subpixels. The new metrics will be useful in the development of new image processing technology and display technology for better image quality.