EnvGS++: Modeling View-Dependency with Real-Time Environment Gaussians

Reconstructing complex reflections in real-world scenes from 2D images is essential for achieving photorealistic novel view synthesis. Existing methods that utilize environment maps to model reflections from distant lighting often struggle with high-frequency reflection details and fail to account for near-field reflections. In this work, we introduce EnvGS++, a novel approach that employs a set of Gaussian primitives as an explicit 3D representation for capturing reflections of environments. EnvGS++ is the journal extension of our preliminary conference work EnvGS [1]. These environment Gaussian primitives are incorporated with base Gaussian primitives to model the appearance of the whole scene. To efficiently render these environment Gaussian primitives, we developed a ray-tracing-based renderer that leverages the GPU's RT core for fast rendering. This allows us to jointly optimize our model for high-quality reconstruction while maintaining real-time rendering speeds. To further boost the rendering speed, we introduce the SnugProxy acceleration strategy for reducing wasted computation in both the rasterization and ray tracing process. Additionally, we propose diffuse regularization to further stabilize and accelerate optimization. Results from multiple real-world and synthetic datasets demonstrate that our method produces significantly more detailed reflections, achieving the best rendering quality in real-time novel view synthesis.