RGBLab

RGBLab is a research laboratory focused on color science, digital imaging, and color management. The group studies how colors are captured, represented, displayed, and perceived across digital systems, with emphasis on accuracy, perceptual relevance, and reproducibility. Core research areas include RGB color spaces and calibration, color appearance models, perceptual color difference metrics, and the engineering of color pipelines for cameras, displays, and printers. The lab also investigates adaptive color processing under varying lighting and the role of machine learning in color reproduction.

Founded in 2010 by a cross-disciplinary team from computer vision, optics, and vision science, RGBLab operates within a university research ecosystem and collaborates with industry partners and standards bodies. It maintains an active portfolio of projects, including open-source software for color management, a publicly accessible Color Patch Dataset for calibration and evaluation, and benchmarks for perceptual color difference.

Facilities include a spectrophotometer, an integrating sphere, calibrated display rigs, and a high-performance computing cluster to support experiments and simulations. The lab publishes findings in peer-reviewed venues and contributes to international standards and practices in color management and imaging pipelines.

RGBLab maintains open collaboration channels, hosting visiting researchers, tutorials, and open data releases to advance the understanding of color representation in digital media and to support faithful color reproduction across devices.