CVJun 16, 2025

GS-2DGS: Geometrically Supervised 2DGS for Reflective Object Reconstruction

arXiv:2506.13110v116 citationsh-index: 8Has CodeCVPR
Originality Incremental advance
AI Analysis

This addresses the problem of efficient and accurate reconstruction of reflective objects for applications in computer vision and graphics, representing an incremental improvement by bridging gaps between existing methods.

The paper tackles the problem of 3D modeling highly reflective objects, which is challenging due to view-dependent appearances, by proposing GS-2DGS, a method that combines 2D Gaussian Splatting with geometric supervision from foundation models. It achieves performance comparable to SDF-based methods while being an order of magnitude faster, as demonstrated on synthetic and real datasets.

3D modeling of highly reflective objects remains challenging due to strong view-dependent appearances. While previous SDF-based methods can recover high-quality meshes, they are often time-consuming and tend to produce over-smoothed surfaces. In contrast, 3D Gaussian Splatting (3DGS) offers the advantage of high speed and detailed real-time rendering, but extracting surfaces from the Gaussians can be noisy due to the lack of geometric constraints. To bridge the gap between these approaches, we propose a novel reconstruction method called GS-2DGS for reflective objects based on 2D Gaussian Splatting (2DGS). Our approach combines the rapid rendering capabilities of Gaussian Splatting with additional geometric information from foundation models. Experimental results on synthetic and real datasets demonstrate that our method significantly outperforms Gaussian-based techniques in terms of reconstruction and relighting and achieves performance comparable to SDF-based methods while being an order of magnitude faster. Code is available at https://github.com/hirotong/GS2DGS

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes