SurfaceSplat: Connecting Surface Reconstruction and Gaussian Splatting

TL;DR

SurfaceSplat introduces a hybrid approach combining SDF and 3D Gaussian Splatting to improve surface reconstruction and novel view rendering from sparse images, outperforming existing methods on benchmark datasets.

Contribution

A novel hybrid method that integrates SDF and 3D Gaussian Splatting, leveraging their strengths for enhanced surface reconstruction and view synthesis.

Findings

Outperforms state-of-the-art on DTU dataset

Improves surface detail and geometry coherence

Effective in sparse-view scenarios

Abstract

Surface reconstruction and novel view rendering from sparse-view images are challenging. Signed Distance Function (SDF)-based methods struggle with fine details, while 3D Gaussian Splatting (3DGS)-based approaches lack global geometry coherence. We propose a novel hybrid method that combines the strengths of both approaches: SDF captures coarse geometry to enhance 3DGS-based rendering, while newly rendered images from 3DGS refine the details of SDF for accurate surface reconstruction. As a result, our method surpasses state-of-the-art approaches in surface reconstruction and novel view synthesis on the DTU and MobileBrick datasets. Code will be released at https://github.com/aim-uofa/SurfaceSplat.