Ref-DGS: Reflective Dual Gaussian Splatting
This work provides a more efficient and accurate method for novel view synthesis and surface reconstruction in highly reflective environments, which is a common challenge for computer graphics and vision researchers.
This paper addresses the challenge of accurately reconstructing surfaces and synthesizing novel views in scenes with strong near-field specular reflections. The authors introduce Ref-DGS, a dual Gaussian splatting framework that decouples surface reconstruction from specular reflection, achieving state-of-the-art performance on reflective scenes while training significantly faster than ray-based methods.
Reflective appearance, especially strong and typically near-field specular reflections, poses a fundamental challenge for accurate surface reconstruction and novel view synthesis. Existing Gaussian splatting methods either fail to model near-field specular reflections or rely on explicit ray tracing at substantial computational cost. We present Ref-DGS, a reflective dual Gaussian splatting framework that addresses this trade-off by decoupling surface reconstruction from specular reflection within an efficient rasterization-based pipeline. Ref-DGS introduces a dual Gaussian scene representation consisting of geometry Gaussians and complementary local reflection Gaussians that capture near-field specular interactions without explicit ray tracing, along with a global environment reflection field for modeling far-field specular reflections. To predict specular radiance, we further propose a lightweight, physically-aware adaptive mixing shader that fuses global and local reflection features. Experiments demonstrate that Ref-DGS achieves state-of-the-art performance on reflective scenes while training substantially faster than ray-based Gaussian methods.