A Case Study in Bridging Art and Technology

The color of oil painting is subjective, and the existing tools are limited in modeling, style adaptation, and detail restoration. Museums and art institutions are faced with the dual challenges of lack of scientific rigor and artistic distortion. In this study, an analysis and simulation framework integrating Commission Internationale de l'Éclairage 1976 L*a*b* color space; hue, saturation, and value color space; and lightweight deep network was constructed, and dynamic feature fusion and closed-loop feedback mechanism were introduced to support high-fidelity reconstruction, cross-genre migration, and real-time interaction. The system was deployed in art galleries and universities, covering more than 120 multigenre oil paintings. The results showed that the color reproduction accuracy was high and the style similarity was 95%. Modular design reduced data dependence, and physical apriority enhanced artistic authenticity. This study provides a reference for interdisciplinary collaboration and scenario verification in digital humanities.