GSFusion: Online RGB-D Mapping Where Gaussian Splatting Meets TSDF Fusion

TL;DR

GSFusion introduces a real-time 3D mapping method that combines Gaussian splatting with volumetric TSDF, reducing artifacts and computational load while improving visualization quality in RGB-D mapping.

Contribution

It integrates Gaussian splatting into a volumetric mapping system using a quadtree structure to enable efficient, high-quality 3D reconstruction in real-time.

Findings

Reduces Gaussian parameters for faster optimization.

Produces high-quality, artifact-free 3D maps.

Demonstrates effectiveness on synthetic and real datasets.

Abstract

Traditional volumetric fusion algorithms preserve the spatial structure of 3D scenes, which is beneficial for many tasks in computer vision and robotics. However, they often lack realism in terms of visualization. Emerging 3D Gaussian splatting bridges this gap, but existing Gaussian-based reconstruction methods often suffer from artifacts and inconsistencies with the underlying 3D structure, and struggle with real-time optimization, unable to provide users with immediate feedback in high quality. One of the bottlenecks arises from the massive amount of Gaussian parameters that need to be updated during optimization. Instead of using 3D Gaussian as a standalone map representation, we incorporate it into a volumetric mapping system to take advantage of geometric information and propose to use a quadtree data structure on images to drastically reduce the number of splats initialized. In this way, we simultaneously generate a compact 3D Gaussian map with fewer artifacts and a volumetric map on the fly. Our method, GSFusion, significantly enhances computational efficiency without sacrificing rendering quality, as demonstrated on both synthetic and real datasets. Code will be available at https://github.com/goldoak/GSFusion.