# NGM-SLAM: Gaussian Splatting SLAM with Radiance Field Submap

arXiv: https://arxiv.org/abs/2405.05702 · HTML: https://mlsanity.com/paper/2405.05702/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 May 2024.

## Built on by (most cited first, 8 of 8)

- [MG-SLAM: Structure Gaussian Splatting SLAM with Manhattan World Hypothesis](https://mlsanity.com/paper/2405.20031/) (2024)
- [MCN-SLAM: Multi-Agent Collaborative Neural SLAM with Hybrid Implicit Neural Scene Representation](https://mlsanity.com/paper/2506.18678/) (2025)
- [MGS-SLAM: Monocular Sparse Tracking and Gaussian Mapping with Depth Smooth Regularization](https://mlsanity.com/paper/2405.06241/) (2024)
- [Multi-Modal UAV Detection, Classification and Tracking Algorithm -- Technical Report for CVPR 2024 UG2 Challenge](https://mlsanity.com/paper/2405.16464/) (2024)
- [LiHi-GS: LiDAR-Supervised Gaussian Splatting for Highway Driving Scene Reconstruction](https://mlsanity.com/paper/2412.15447/) (2024)
- [LLGS: Unsupervised Gaussian Splatting for Image Enhancement and Reconstruction in Pure Dark Environment](https://mlsanity.com/paper/2503.18640/) (2025)
- [OG-Mapping: Octree-based Structured 3D Gaussians for Online Dense Mapping](https://mlsanity.com/paper/2408.17223/) (2024)
- [Towards Next-Generation SLAM: A Survey on 3DGS-SLAM Focusing on Performance, Robustness, and Future Directions](https://mlsanity.com/paper/2602.04251/) (2026)

## Abstract (arXiv)

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.