MG-SLAM: Structure Gaussian Splatting SLAM with Manhattan World Hypothesis

2024-05-30

AI Summary

To address severe incompleteness and voids in Gaussian Splatting SLAM reconstructions within complex indoor scenes, this paper proposes the first structure-aware Gaussian Splatting SLAM framework incorporating the Manhattan world assumption. Methodologically, it jointly leverages line segment detection and planar priors to impose structured geometric constraints on Gaussian optimization and interpolation, enabling intelligent geometric completion in texture-poor or occluded regions; tracking and mapping are co-optimized in real time. The core contribution is the first integration of Manhattan structural priors into Gaussian Splatting SLAM, achieved via synergistic line–plane modeling to enhance geometric consistency and robustness. Evaluated on both synthetic and real-world datasets, the method achieves state-of-the-art performance: voids are significantly reduced, reconstruction completeness and geometric accuracy are markedly improved, while maintaining real-time operation.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Intelligent Robots: State Estimation
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Gaussian Splatting SLAMs have made significant advancements in improving the efficiency and fidelity of real-time reconstructions. However, these systems often encounter incomplete reconstructions in complex indoor environments, characterized by substantial holes due to unobserved geometry caused by obstacles or limited view angles. To address this challenge, we present Manhattan Gaussian SLAM, an RGB-D system that leverages the Manhattan World hypothesis to enhance geometric accuracy and completeness. By seamlessly integrating fused line segments derived from structured scenes, our method ensures robust tracking in textureless indoor areas. Moreover, The extracted lines and planar surface assumption allow strategic interpolation of new Gaussians in regions of missing geometry, enabling efficient scene completion. Extensive experiments conducted on both synthetic and real-world scenes demonstrate that these advancements enable our method to achieve state-of-the-art performance, marking a substantial improvement in the capabilities of Gaussian SLAM systems.

Problem

• Improves geometric accuracy in complex indoor environments
• Enhances scene completeness using Manhattan World hypothesis
• Enables robust tracking in textureless indoor areas

Innovation

• Integrates Manhattan World hypothesis for accuracy
• Uses fused line segments for robust tracking
• Interpolates Gaussians for efficient scene completion