FGS-SLAM: Fourier-based Gaussian Splatting for Real-time SLAM with Sparse and Dense Map Fusion

2025-03-03

AI Summary

In 3D Gaussian Splatting SLAM, slow convergence and geometric representation redundancy or insufficiency arise from uncertain Gaussian initialization. To address this, we propose a Fourier frequency-domain analysis–based adaptive densification method to construct geometry-aware Gaussian priors. We further design a sparse-dense dual-map collaborative architecture: a sparse map enables efficient GICP-based tracking, while a unified dense map ensures high-fidelity visual reconstruction. This work is the first to introduce frequency-domain analysis into Gaussian SLAM initialization and to establish a novel functional decoupling and fusion mechanism between sparse and dense maps. Evaluated on the Replica and TUM RGB-D benchmarks, our system achieves real-time performance at 36 FPS, with both localization and mapping accuracy surpassing state-of-the-art methods.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Computer Vision: Motion & Tracking
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Security and Privacy: Large-scale security measurements

Abstract

3D gaussian splatting has advanced simultaneous localization and mapping (SLAM) technology by enabling real-time positioning and the construction of high-fidelity maps. However, the uncertainty in gaussian position and initialization parameters introduces challenges, often requiring extensive iterative convergence and resulting in redundant or insufficient gaussian representations. To address this, we introduce a novel adaptive densification method based on Fourier frequency domain analysis to establish gaussian priors for rapid convergence. Additionally, we propose constructing independent and unified sparse and dense maps, where a sparse map supports efficient tracking via Generalized Iterative Closest Point (GICP) and a dense map creates high-fidelity visual representations. This is the first SLAM system leveraging frequency domain analysis to achieve high-quality gaussian mapping in real-time. Experimental results demonstrate an average frame rate of 36 FPS on Replica and TUM RGB-D datasets, achieving competitive accuracy in both localization and mapping.

Problem

• Addresses uncertainty in Gaussian position and initialization parameters.
• Proposes adaptive densification using Fourier frequency domain analysis.
• Develops independent sparse and dense maps for efficient SLAM.

Innovation

• Fourier-based adaptive densification for rapid convergence
• Independent sparse and dense map fusion
• Real-time SLAM with frequency domain analysis