Xiaolei Lang
Jiajun Lv
Kai Tang
Laijian Li
Jianxin Huang
Lina Liu
Yong Liu
Xingxing Zuo
This paper proposes an innovative LiDAR-Inertial-Camera SLAM system with 3D Gaussian Splatting, which is the first to jointly consider visual quality, geometric accuracy, and real-time performance. It robustly and accurately estimates poses while building a photo-realistic 3D Gaussian map in real time that enables high-quality novel view RGB and depth rendering. To effectively address under-reconstruction in regions not covered by the LiDAR, we employ a lightweight zero-shot depth model that synergistically combines RGB appearance cues with sparse LiDAR measurements to generate dense depth maps. The depth completion enables reliable Gaussian initialization in LiDAR-blind areas, significantly improving system applicability for sparse LiDAR sensors. To enhance geometric accuracy, we use sparse but precise LiDAR depths to supervise Gaussian map optimization and accelerate it with carefully designed CUDA-accelerated strategies. Furthermore, we explore how the incrementally reconstructed Gaussian map can improve the robustness of odometry. By tightly incorporating photometric constraints from the Gaussian map into the continuous-time factor graph optimization, we demonstrate improved pose estimation under LiDAR degradation scenarios. We also showcase downstream applications via extending our elaborate system, including video frame interpolation and fast 3D mesh extraction. To support rigorous evaluation, we construct a dedicated LiDAR-Inertial-Camera dataset featuring ground-truth poses, depth maps, and extrapolated trajectories for assessing out-of-sequence novel view synthesis. Extensive experiments on both public and self-collected datasets demonstrate the superiority and versatility of our system across LiDAR sensors with varying sampling densities. Both the dataset and code will be made publicly available on project page https://xingxingzuo.github.io/gaussian_lic2.
PDF URL