DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization

TL;DR

DG-SLAM introduces a robust dynamic visual SLAM system using 3D Gaussian models, achieving high accuracy in pose estimation and map reconstruction in dynamic scenes while maintaining real-time performance.

Contribution

It is the first SLAM system based on 3D Gaussians that effectively handles dynamic environments with novel strategies for motion masking and hybrid tracking.

Findings

Outperforms existing methods in pose accuracy and map quality

Maintains real-time rendering in dynamic scenes

Provides high-fidelity reconstructions and novel-view synthesis

Abstract

Achieving robust and precise pose estimation in dynamic scenes is a significant research challenge in Visual Simultaneous Localization and Mapping (SLAM). Recent advancements integrating Gaussian Splatting into SLAM systems have proven effective in creating high-quality renderings using explicit 3D Gaussian models, significantly improving environmental reconstruction fidelity. However, these approaches depend on a static environment assumption and face challenges in dynamic environments due to inconsistent observations of geometry and photometry. To address this problem, we propose DG-SLAM, the first robust dynamic visual SLAM system grounded in 3D Gaussians, which provides precise camera pose estimation alongside high-fidelity reconstructions. Specifically, we propose effective strategies, including motion mask generation, adaptive Gaussian point management, and a hybrid camera tracking algorithm to improve the accuracy and robustness of pose estimation. Extensive experiments demonstrate that DG-SLAM delivers state-of-the-art performance in camera pose estimation, map reconstruction, and novel-view synthesis in dynamic scenes, outperforming existing methods meanwhile preserving real-time rendering ability.