Dense Point Clouds Matter: Dust-GS for Scene Reconstruction from Sparse Viewpoints
This addresses a domain-specific bottleneck in computer vision for applications like robotics or AR/VR where capturing many images is impractical, though it appears incremental as it builds on 3DGS.
The paper tackles the problem of 3D scene reconstruction from sparse viewpoints, where traditional 3D Gaussian Splatting (3DGS) struggles due to poor initial point clouds, and presents Dust-GS, which achieves superior reconstruction quality with fewer input images on benchmark datasets.
3D Gaussian Splatting (3DGS) has demonstrated remarkable performance in scene synthesis and novel view synthesis tasks. Typically, the initialization of 3D Gaussian primitives relies on point clouds derived from Structure-from-Motion (SfM) methods. However, in scenarios requiring scene reconstruction from sparse viewpoints, the effectiveness of 3DGS is significantly constrained by the quality of these initial point clouds and the limited number of input images. In this study, we present Dust-GS, a novel framework specifically designed to overcome the limitations of 3DGS in sparse viewpoint conditions. Instead of relying solely on SfM, Dust-GS introduces an innovative point cloud initialization technique that remains effective even with sparse input data. Our approach leverages a hybrid strategy that integrates an adaptive depth-based masking technique, thereby enhancing the accuracy and detail of reconstructed scenes. Extensive experiments conducted on several benchmark datasets demonstrate that Dust-GS surpasses traditional 3DGS methods in scenarios with sparse viewpoints, achieving superior scene reconstruction quality with a reduced number of input images.