MCGS-SLAM: A Multi-Camera SLAM Framework Using Gaussian Splatting for High-Fidelity Mapping

2025-09-17

AI Summary

To address the robustness and geometric coverage limitations of monocular SLAM—particularly in lateral regions—this paper proposes the first purely RGB multi-camera SLAM system, integrating 3D Gaussian splatting with dense multi-view visual information for high-fidelity mapping and accurate trajectory estimation. Methodologically, it introduces (1) multi-camera geometric constraints into the 3D Gaussian optimization framework; (2) a multi-camera bundle adjustment (MCBA) formulation that jointly optimizes camera poses and Gaussian parameters; and (3) a low-rank prior-driven scale-consistency module ensuring metric alignment across views. The system operates in real time at large scale and significantly outperforms monocular baselines on both synthetic and real-world datasets. It successfully reconstructs lateral structures missed by monocular systems and achieves photorealistic reconstruction quality.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Recent progress in dense SLAM has primarily targeted monocular setups, often at the expense of robustness and geometric coverage. We present MCGS-SLAM, the first purely RGB-based multi-camera SLAM system built on 3D Gaussian Splatting (3DGS). Unlike prior methods relying on sparse maps or inertial data, MCGS-SLAM fuses dense RGB inputs from multiple viewpoints into a unified, continuously optimized Gaussian map. A multi-camera bundle adjustment (MCBA) jointly refines poses and depths via dense photometric and geometric residuals, while a scale consistency module enforces metric alignment across views using low-rank priors. The system supports RGB input and maintains real-time performance at large scale. Experiments on synthetic and real-world datasets show that MCGS-SLAM consistently yields accurate trajectories and photorealistic reconstructions, usually outperforming monocular baselines. Notably, the wide field of view from multi-camera input enables reconstruction of side-view regions that monocular setups miss, critical for safe autonomous operation. These results highlight the promise of multi-camera Gaussian Splatting SLAM for high-fidelity mapping in robotics and autonomous driving.

Problem

• Develops multi-camera SLAM using Gaussian splatting for high-fidelity mapping
• Addresses monocular SLAM limitations in robustness and geometric coverage
• Enables reconstruction of side-view regions critical for autonomous operation

Innovation

• Multi-camera Gaussian Splatting for dense mapping
• Joint pose-depth optimization via photometric residuals
• Scale consistency with low-rank metric alignment