Optimizing 3D Gaussian Splatting for Sparse Viewpoint Scene Reconstruction

TL;DR

This paper introduces SVS-GS, a novel framework that enhances 3D Gaussian Splatting for sparse viewpoint scene reconstruction by reducing artifacts and improving geometric consistency, thus broadening its practical applicability.

Contribution

The paper proposes SVS-GS, integrating a Gaussian smoothing filter, DGPP loss with a depth mask, and SDS loss to improve sparse viewpoint 3D scene reconstruction.

Findings

Significant reduction in artifacts in sparse view reconstructions.

Improved geometric consistency and edge sharpness.

Enhanced performance on MipNeRF-360 and SeaThru-NeRF datasets.

Abstract

3D Gaussian Splatting (3DGS) has emerged as a promising approach for 3D scene representation, offering a reduction in computational overhead compared to Neural Radiance Fields (NeRF). However, 3DGS is susceptible to high-frequency artifacts and demonstrates suboptimal performance under sparse viewpoint conditions, thereby limiting its applicability in robotics and computer vision. To address these limitations, we introduce SVS-GS, a novel framework for Sparse Viewpoint Scene reconstruction that integrates a 3D Gaussian smoothing filter to suppress artifacts. Furthermore, our approach incorporates a Depth Gradient Profile Prior (DGPP) loss with a dynamic depth mask to sharpen edges and 2D diffusion with Score Distillation Sampling (SDS) loss to enhance geometric consistency in novel view synthesis. Experimental evaluations on the MipNeRF-360 and SeaThru-NeRF datasets demonstrate that SVS-GS markedly improves 3D reconstruction from sparse viewpoints, offering a robust and efficient solution for scene understanding in robotics and computer vision applications.