White Balance

White balance (WB) correction aims to neutralize color casts in images caused by varying light sources, ensuring accurate color reproduction. Current research focuses on improving the speed and accuracy of automatic white balancing (AWB) algorithms, employing deep learning models such as 3D lookup tables (LUTs) and attention-based networks to handle both single and multiple illuminants. These advancements leverage techniques like contrastive learning and deterministic illumination mapping to achieve real-time performance and superior color accuracy compared to traditional methods. The resulting improvements in image quality have significant implications for various applications, including photography, computer vision, and underwater imaging.