FGO-SLAM: Enhancing Gaussian SLAM with Globally Consistent Opacity Radiance Field

2025-09-01

AI Summary

To address insufficient pose optimization and low geometric reconstruction accuracy in Gaussian SLAM, this paper proposes an opacity-field-based Gaussian SLAM system. Methodologically, it introduces, for the first time, a globally consistent opacity radiance field into 3D Gaussian representations, enabling explicit surface extraction directly from Gaussian distributions. It jointly optimizes camera poses and point clouds, incorporating depth distortion constraints and normal consistency regularization to enhance geometric fidelity. Furthermore, it integrates pose graph optimization, sparse point cloud refinement, tetrahedral meshing, and level-set surface reconstruction. Evaluated on multiple real-world and large-scale synthetic datasets, the method achieves significant improvements in tracking accuracy and dense mapping quality. It attains state-of-the-art performance in geometric completeness, surface detail preservation, and system robustness.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Computer Vision: 3D Computer Vision

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: Experiences and lessons learnt from Web-based algorithms and system deployments
• Security and Privacy: Data transparency and provenance

Abstract

Visual SLAM has regained attention due to its ability to provide perceptual capabilities and simulation test data for Embodied AI. However, traditional SLAM methods struggle to meet the demands of high-quality scene reconstruction, and Gaussian SLAM systems, despite their rapid rendering and high-quality mapping capabilities, lack effective pose optimization methods and face challenges in geometric reconstruction. To address these issues, we introduce FGO-SLAM, a Gaussian SLAM system that employs an opacity radiance field as the scene representation to enhance geometric mapping performance. After initial pose estimation, we apply global adjustment to optimize camera poses and sparse point cloud, ensuring robust tracking of our approach. Additionally, we maintain a globally consistent opacity radiance field based on 3D Gaussians and introduce depth distortion and normal consistency terms to refine the scene representation. Furthermore, after constructing tetrahedral grids, we identify level sets to directly extract surfaces from 3D Gaussians. Results across various real-world and large-scale synthetic datasets demonstrate that our method achieves state-of-the-art tracking accuracy and mapping performance.

Problem

• Enhancing geometric reconstruction in Gaussian SLAM systems
• Optimizing camera poses and sparse point clouds globally
• Extracting surfaces directly from 3D Gaussians using level sets

Innovation

• Opacity radiance field enhances geometric mapping
• Global adjustment optimizes camera poses and point cloud
• Depth distortion and normal terms refine scene representation