NGM-SLAM: Gaussian Splatting SLAM with Radiance Field Submap

TL;DR

NGM-SLAM introduces a novel SLAM system combining neural radiance field submaps with Gaussian Splatting, enabling high-quality, real-time scene reconstruction and effective loop closure in large-scale environments.

Contribution

It is the first 3D Gaussian Splatting SLAM system that uses neural radiance field submaps for progressive scene expression and improved loop closure.

Findings

Achieves state-of-the-art scene reconstruction and tracking.

Supports monocular, stereo, and RGB-D inputs.

Demonstrates high-quality scene expression and hole filling.

Abstract

SLAM systems based on Gaussian Splatting have garnered attention due to their capabilities for rapid real-time rendering and high-fidelity mapping. However, current Gaussian Splatting SLAM systems usually struggle with large scene representation and lack effective loop closure detection. To address these issues, we introduce NGM-SLAM, the first 3DGS based SLAM system that utilizes neural radiance field submaps for progressive scene expression, effectively integrating the strengths of neural radiance fields and 3D Gaussian Splatting. We utilize neural radiance field submaps as supervision and achieve high-quality scene expression and online loop closure adjustments through Gaussian rendering of fused submaps. Our results on multiple real-world scenes and large-scale scene datasets demonstrate that our method can achieve accurate hole filling and high-quality scene expression, supporting monocular, stereo, and RGB-D inputs, and achieving state-of-the-art scene reconstruction and tracking performance.