MotionGS : Compact Gaussian Splatting SLAM by Motion Filter

TL;DR

MotionGS introduces a compact 3D Gaussian splatting SLAM method that combines feature-based tracking, motion filtering, and dual keyframe selection, achieving improved accuracy and efficiency over existing SLAM approaches.

Contribution

The paper presents a novel 3D Gaussian splatting SLAM approach integrating deep visual features, motion filtering, and dual keyframe selection for enhanced performance and reduced memory usage.

Findings

Outperforms existing SLAM methods in tracking accuracy

Achieves more efficient mapping with less memory consumption

Provides robust coarse-to-fine pose estimation

Abstract

With their high-fidelity scene representation capability, the attention of SLAM field is deeply attracted by the Neural Radiation Field (NeRF) and 3D Gaussian Splatting (3DGS). Recently, there has been a surge in NeRF-based SLAM, while 3DGS-based SLAM is sparse. A novel 3DGS-based SLAM approach with a fusion of deep visual feature, dual keyframe selection and 3DGS is presented in this paper. Compared with the existing methods, the proposed tracking is achieved by feature extraction and motion filter on each frame. The joint optimization of poses and 3D Gaussians runs through the entire mapping process. Additionally, the coarse-to-fine pose estimation and compact Gaussian scene representation are implemented by dual keyframe selection and novel loss functions. Experimental results demonstrate that the proposed algorithm not only outperforms the existing methods in tracking and mapping, but also has less memory usage.