Real-time light calculation and optimization model for large-scale scene rendering

With the rapid development of virtual reality, digital twins and large-scale game scenes, how to realize real-time rendering of large-scale complex scenes while maintaining visual realism has become a core issue in computer graphics research. As a key link affecting the rendering quality and performance, lighting computation is often difficult to balance the efficiency and quality of traditional methods in the face of massive geometric data, dynamic light sources, and variable environments. In this paper, we propose a real-time lighting computation and optimization model for large-scale scenes. By constructing a hierarchical lighting structure, introducing efficient light sampling and masking algorithms, and combining with GPU parallel computation optimization strategies, we achieve both high-quality lighting effects and real-time performance in complex scenes. The experimental results show that the model shows good stability, rendering efficiency and visual consistency in several typical large-scale rendering scenes, and has significant advantages in performance and image quality compared with the existing mainstream methods, which provides effective support and theoretical basis for future real-time rendering systems with larger scale and more dynamic elements.