LVD-GS: Gaussian Splatting SLAM for Dynamic Scenes via Hierarchical Explicit-Implicit Representation Collaboration Rendering

TL;DR

LVD-GS introduces a hierarchical collaborative representation and dynamic object modeling in LiDAR-Visual Gaussian Splatting SLAM, significantly improving large-scale dynamic scene mapping accuracy and robustness.

Contribution

It presents a novel hierarchical collaboration module and dynamic modeling approach to enhance SLAM performance in dynamic outdoor environments.

Findings

Achieves state-of-the-art results on KITTI, nuScenes, and new datasets.

Effectively mitigates scale drift and enhances reconstruction robustness.

Accurately detects and models dynamic objects in complex scenes.

Abstract

3D Gaussian Splatting SLAM has emerged as a widely used technique for high-fidelity mapping in spatial intelligence. However, existing methods often rely on a single representation scheme, which limits their performance in large-scale dynamic outdoor scenes and leads to cumulative pose errors and scale ambiguity. To address these challenges, we propose \textbf{LVD-GS}, a novel LiDAR-Visual 3D Gaussian Splatting SLAM system. Motivated by the human chain-of-thought process for information seeking, we introduce a hierarchical collaborative representation module that facilitates mutual reinforcement for mapping optimization, effectively mitigating scale drift and enhancing reconstruction robustness. Furthermore, to effectively eliminate the influence of dynamic objects, we propose a joint dynamic modeling module that generates fine-grained dynamic masks by fusing open-world segmentation with implicit residual constraints, guided by uncertainty estimates from DINO-Depth features. Extensive evaluations on KITTI, nuScenes, and self-collected datasets demonstrate that our approach achieves state-of-the-art performance compared to existing methods.