MGS-SLAM: Monocular Sparse Tracking and Gaussian Mapping with Depth Smooth Regularization
This addresses geometric inaccuracies in monocular SLAM systems for robotics and AR/VR applications, representing an incremental improvement over existing Gaussian Splatting approaches.
The paper tackles the problem of inaccurate geometric reconstruction and weak tracking in monocular Gaussian Splatting SLAM by jointly optimizing sparse visual odometry tracking and 3D Gaussian Splatting scene representation, achieving state-of-the-art pose estimation accuracy and outperforming previous monocular methods in novel view synthesis and geometric reconstruction.
This letter introduces a novel framework for dense Visual Simultaneous Localization and Mapping (VSLAM) based on Gaussian Splatting. Recently, SLAM based on Gaussian Splatting has shown promising results. However, in monocular scenarios, the Gaussian maps reconstructed lack geometric accuracy and exhibit weaker tracking capability. To address these limitations, we jointly optimize sparse visual odometry tracking and 3D Gaussian Splatting scene representation for the first time. We obtain depth maps on visual odometry keyframe windows using a fast Multi-View Stereo (MVS) network for the geometric supervision of Gaussian maps. Furthermore, we propose a depth smooth loss and Sparse-Dense Adjustment Ring (SDAR) to reduce the negative effect of estimated depth maps and preserve the consistency in scale between the visual odometry and Gaussian maps. We have evaluated our system across various synthetic and real-world datasets. The accuracy of our pose estimation surpasses existing methods and achieves state-of-the-art. Additionally, it outperforms previous monocular methods in terms of novel view synthesis and geometric reconstruction fidelities.