Search papers, labs, and topics across Lattice.
This paper introduces CGS-SLAM, a hybrid decentralized/centralized system that enables multi-agent 3D Gaussian Splatting (3DGS) simultaneous localization and mapping (SLAM) using only RGB and inertial data, addressing the limitations of existing methods that rely on RGB-D input. By employing local tracking with inertial data and a monocular depth estimator, the system allows agents to share keyframe encodings and dynamically align submaps in overlapping regions, which enhances overall mapping accuracy. Experimental results show that CGS-SLAM achieves competitive tracking performance and improved rendering quality in challenging GNSS-denied environments compared to state-of-the-art approaches.
Multi-agent SLAM can now achieve high-quality 3D reconstruction using only RGB and inertial data, making it accessible for consumer-grade devices.
Recent advances in SLAM have leveraged 3DGS for photorealistic reconstruction and novel view synthesis. However, most methods rely on RGB-D input, which is unavailable on consumer-grade smartphones, and few integrate 3DGS within a collaborative framework. Therefore, we present CGS-SLAM, a hybrid decentralized/centralized system enabling multi-agent 3DGS SLAM using only RGB and inertial data. Each agent performs local tracking with inertial data as a motion prior and reconstructs a scaled map using a metric monocular depth estimator (Depth Pro). Keyframe encodings are shared among agents, enabling dynamic keyframing in regions of spatial overlaps with other agents, enhancing submap alignment. Afterwards, a central server aligns submaps using VGGT as a view alignment model. This bidirectional communication keeps communication cost low during mapping and global reconstruction in difficult GNSS-denied environments. Experiments on multiple datasets demonstrate competitive tracking performance, improved rendering quality over state-of-the-art methods, and accurate submap alignment.
