SSR-GS: Separating Specular Reflection in Gaussian Splatting for Glossy Surface Reconstruction

TL;DR

SSR-GS, a specular reflection modeling framework for glossy surface reconstruction, is proposed, which introduces a prefiltered Mip-Cubemap to model direct specular reflections efficiently, and proposes an IndiASG module to capture indirect specular reflections.

Abstract

In recent years, 3D Gaussian splatting (3DGS) has achieved remarkable progress in novel view synthesis. However, accurately reconstructing glossy surfaces under complex illumination remains challenging, particularly in scenes with strong specular reflections and multi-surface interreflections. To address this issue, we propose SSR-GS, a specular reflection modeling framework for glossy surface reconstruction. Specifically, we introduce a prefiltered Mip-Cubemap to model direct specular reflections efficiently, and propose an IndiASG module to capture indirect specular reflections. Furthermore, we design Visual Geometry Priors (VGP) that couple a reflection-aware visual prior via a reflection score (RS) to downweight the photometric loss contribution of reflection-dominated regions, with geometry priors derived from VGGT, including progressively decayed depth supervision and transformed normal constraints. Extensive experiments on both synthetic and real-world datasets demonstrate that SSR-GS achieves state-of-the-art performance in glossy surface reconstruction.

Figures (5)

• Figure 1: Surface reconstruction results on toaster. CD denotes the Chamfer distance.
• Figure 2: An overview of our SSR-GS pipeline. We rasterize 3D Gaussians to obtain surface buffers including normals, depth, opacity, diffuse component $C_{\mathrm{diff}}$, roughness, $F_0$, and alpha, and supervise them with geometry priors (GP). After rasterization, we extract a mesh via TSDF fusion and perform mesh-based ray tracing to estimate visibility $w_{\mathrm{vis}}$. Direct specular reflection is queried from a Mip-Cubemap environment map, while indirect specular reflection is modeled by IndiASG. Finally, we apply physically based deferred rendering (Eq. \ref{['eq:final_color']}) to produce the rendered image. We additionally compute the visual prior (VP) (reflection score) and use it to down-weight reflection-dominated pixels in the photometric loss for stable geometry initialization.
• Figure 3: Qualitative results of surface reconstruction on ShinySynthetic (car, coffee) and GlossySynthetic (cat, bell, teapot) datasets. Since the released GS-ROR$^2$ code cannot extract meshes on the ShinySynthetic, we instead visualize its normal results for comparison.
• Figure 4: Qualitative results of surface reconstruction on Ref-Real (gardenspheres, toycar) dataset.
• Figure 5: Ablation study on different components.