GCC: Generative Color Constancy via Diffusing a Color Checker
This addresses the issue of camera sensor variability in color constancy for real-world applications, representing a novel method for a known bottleneck.
The paper tackles the problem of color constancy methods struggling to generalize across different camera sensors by proposing GCC, which uses diffusion models to inpaint color checkers into images for illumination estimation, achieving strong robustness in cross-camera scenarios without sensor-specific training.
Color constancy methods often struggle to generalize across different camera sensors due to varying spectral sensitivities. We present GCC, which leverages diffusion models to inpaint color checkers into images for illumination estimation. Our key innovations include (1) a single-step deterministic inference approach that inpaints color checkers reflecting scene illumination, (2) a Laplacian decomposition technique that preserves checker structure while allowing illumination-dependent color adaptation, and (3) a mask-based data augmentation strategy for handling imprecise color checker annotations. By harnessing rich priors from pre-trained diffusion models, GCC demonstrates strong robustness in challenging cross-camera scenarios. These results highlight our method's effective generalization capability across different camera characteristics without requiring sensor-specific training, making it a versatile and practical solution for real-world applications.