LV-GS SLAM: A Decoupled LiDAR–Visual 3D Gaussian Splatting SLAM in a Large-Scale Environment
Open Access
- 12 August 2026
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 16 (16) , 8028
Abstract
3D Gaussian Splatting (3DGS) has gained prominence in autonomous driving and robotics for its rendering efficiency and high-fidelity reconstruction capabilities. However, incremental 3DGS map construction at large scales remains challenging due to sensor sparsity and computational constraints. In this paper, we propose LV-GS SLAM, a novel system that integrates LiDAR and visual data for incremental, large-scale reconstruction with real-time tracking. This system provides fast and robust pose estimation while also enabling photorealistic rendering. We first employ a LiDAR odometry frontend that processes 30 Hz LiDAR inputs to provide robust initial poses. In our implementation, the complete LiDAR tracking pipeline runs at 15–19 Hz, while the mapping module performs incremental optimization on selected keyframes. To address the sparsity-induced surface discontinuity in conventional LiDAR-based reconstruction, we propose a novel depth propagation approach that initializes 3D Gaussian primitives using dense depth maps, achieving faster PSNR convergence with 9× fewer optimization iterations compared to direct LiDAR initialization. Furthermore, we develop a keyframe-based submap management framework that dynamically adjusts memory allocation based on both primitive density and inter-frame overlap ratio, effectively preventing GPU memory overflow. Our system has been validated on the KITTI dataset, achieving superior rendering quality compared with representative reproducible baselines. We further validate the robustness of the system on a quadruped robot platform, demonstrating satisfactory performance in both pose estimation and high-fidelity reconstruction.Keywords
This publication has 47 references indexed in Scilit:
- RID-LIO: robust and accurate intensity-assisted LiDAR-based SLAM for degenerated environmentsMeasurement Science and Technology, 2025
- CG-SLAM: Efficient Dense RGB-D SLAM in a Consistent Uncertainty-Aware 3D Gaussian Field2024
- NICER-SLAM: Neural Implicit Scene Encoding for RGB SLAM2024
- City-scale continual neural semantic mapping with three-layer sampling and panoptic representationKnowledge-Based Systems, 2023
- Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields2022
- NICE-SLAM: Neural Implicit Scalable Encoding for SLAM2022
- pixelNeRF: Neural Radiance Fields from One or Few Images2021
- NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis2020
- Building Multiversal Semantic Maps for Mobile Robot OperationKnowledge-Based Systems, 2017
- Are we ready for autonomous driving? The KITTI vision benchmark suite2012