GGD-SLAM: Monocular 3DGS SLAM Powered by Generalizable Motion Model for Dynamic Environments

TL;DR

GGD-SLAM introduces a novel monocular SLAM framework that effectively handles dynamic environments by employing a generalizable motion model, dynamic feature separation, and occlusion filling, achieving state-of-the-art results.

Contribution

The paper proposes a dynamic SLAM system that does not rely on semantic annotations or depth input, using a motion model and dynamic feature enhancement for improved performance.

Findings

Achieves state-of-the-art camera pose estimation in dynamic scenes.

Effectively separates static and dynamic components for dense mapping.

Improves robustness against dynamic distractors with a novel SSIM loss.

Abstract

Visual SLAM algorithms achieve significant improvements through the exploration of 3D Gaussian Splatting (3DGS) representations, particularly in generating high-fidelity dense maps. However, they depend on a static environment assumption and experience significant performance degradation in dynamic environments. This paper presents GGD-SLAM, a framework that employs a generalizable motion model to address the challenges of localization and dense mapping in dynamic environments - without predefined semantic annotations or depth input. Specifically, the proposed system employs a First-In-First-Out (FIFO) queue to manage incoming frames, facilitating dynamic semantic feature extraction through a sequential attention mechanism. This is integrated with a dynamic feature enhancer to separate static and dynamic components. Additionally, to minimize dynamic distractors' impact on the static components, we devise a method to fill occluded areas via static information sampling and design a distractor-adaptive Structure Similarity Index Measure (SSIM) loss tailored for dynamic environments, significantly enhancing the system's resilience. Experiments conducted on real-world dynamic datasets demonstrate that the proposed system achieves state-of-the-art performance in camera pose estimation and dense reconstruction in dynamic scenes.