LVI-GS: Tightly Coupled LiDAR–Visual–Inertial SLAM Using 3-D Gaussian Splatting

Abstract

Three-dimensional Gaussian splatting (3DGS) has shown its ability in rapid rendering and high-fidelity mapping. In this article, we introduce a tightly coupled LiDAR-visual–inertial SLAM using 3-D Gaussian splatting (LVI-GS), which leverages the complementary characteristics of light detection and ranging (LiDAR) and image sensors to capture both geometric structures and visual details of 3-D scenes. To this end, the 3-D Gaussians are initialized from colorized LiDAR points and optimized using differentiable rendering. To achieve high-fidelity mapping, we introduce a pyramid-based training approach to effectively learn multilevel features and incorporate depth loss derived from LiDAR measurements to improve geometric feature perception. Through well-designed strategies for Gaussian map expansion, keyframe selection, thread management, and custom compute unified device architecture (CUDA) acceleration, our framework achieves real-time photorealistic mapping. Numerical experiments are performed to evaluate the superior performance of our method compared with state-of-the-art 3-D reconstruction systems. Videos of the evaluations can be found on our website: https://kwanwaipang.github.io/LVI-GS/.