Sparse2DGS: Sparse-View Surface Reconstruction using 2D Gaussian Splatting with Dense Point Cloud

TL;DR

Sparse2DGS introduces a method that combines dense point cloud generation from limited images with Gaussian Splatting for fast, accurate 3D surface reconstruction, overcoming limitations of sparse initializations.

Contribution

It integrates dense point cloud generation with Gaussian Splatting to enable high-quality 3D reconstruction from only three images.

Findings

Accurately reconstructs 3D shapes with just three images

Outperforms existing methods on the DTU dataset

Uses dense point clouds for better Gaussian initialization

Abstract

Gaussian Splatting (GS) has gained attention as a fast and effective method for novel view synthesis. It has also been applied to 3D reconstruction using multi-view images and can achieve fast and accurate 3D reconstruction. However, GS assumes that the input contains a large number of multi-view images, and therefore, the reconstruction accuracy significantly decreases when only a limited number of input images are available. One of the main reasons is the insufficient number of 3D points in the sparse point cloud obtained through Structure from Motion (SfM), which results in a poor initialization for optimizing the Gaussian primitives. We propose a new 3D reconstruction method, called Sparse2DGS, to enhance 2DGS in reconstructing objects using only three images. Sparse2DGS employs DUSt3R, a fundamental model for stereo images, along with COLMAP MVS to generate highly accurate and dense 3D point clouds, which are then used to initialize 2D Gaussians. Through experiments on the DTU dataset, we show that Sparse2DGS can accurately reconstruct the 3D shapes of objects using just three images. The project page is available at https://gsisaoki.github.io/SPARSE2DGS/