Monocular Gaussian SLAM with Language Extended Loop Closure

TL;DR

This paper introduces MG-SLAM, a monocular Gaussian SLAM system that uses 3D Gaussian maps and language-extended loop closure to achieve drift correction, high-fidelity environment reconstruction, and high-level scene understanding.

Contribution

It proposes a novel monocular Gaussian SLAM approach with a language-extended loop closure module based on CLIP features for global optimization.

Findings

Outperforms some RGB-D methods on challenging datasets

Achieves drift correction and high-fidelity mapping

Provides high-level environment understanding

Abstract

Recently,3DGaussianSplattinghasshowngreatpotentialin visual Simultaneous Localization And Mapping (SLAM). Existing methods have achieved encouraging results on RGB-D SLAM, but studies of the monocular case are still scarce. Moreover, they also fail to correct drift errors due to the lack of loop closure and global optimization. In this paper, we present MG-SLAM, a monocular Gaussian SLAM with a language-extended loop closure module capable of performing drift-corrected tracking and high-fidelity reconstruction while achieving a high-level understanding of the environment. Our key idea is to represent the global map as 3D Gaussian and use it to guide the estimation of the scene geometry, thus mitigating the efforts of missing depth information. Further, an additional language-extended loop closure module which is based on CLIP feature is designed to continually perform global optimization to correct drift errors accumulated as the system runs. Our system shows promising results on multiple challenging datasets in both tracking and mapping and even surpasses some existing RGB-D methods.