MotionGS : Compact Gaussian Splatting SLAM by Motion Filter
This work addresses SLAM researchers and practitioners by providing a more compact and efficient 3DGS-based SLAM solution, though it appears incremental as it builds on existing 3DGS and SLAM techniques.
The paper tackles the problem of sparse 3D Gaussian Splatting (3DGS)-based SLAM by proposing MotionGS, which integrates deep visual features, dual keyframe selection, and 3DGS to achieve tracking via feature extraction and motion filtering. The result is a method that outperforms existing approaches in tracking and mapping accuracy while using less memory.
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.