DGS-SLAM: Gaussian Splatting SLAM in Dynamic Environment
Reviewed by Pithpith:3CLHD4LFopen to challenge →
read the original abstract
We introduce Dynamic Gaussian Splatting SLAM (DGS-SLAM), the first dynamic SLAM framework built on the foundation of Gaussian Splatting. While recent advancements in dense SLAM have leveraged Gaussian Splatting to enhance scene representation, most approaches assume a static environment, making them vulnerable to photometric and geometric inconsistencies caused by dynamic objects. To address these challenges, we integrate Gaussian Splatting SLAM with a robust filtering process to handle dynamic objects throughout the entire pipeline, including Gaussian insertion and keyframe selection. Within this framework, to further improve the accuracy of dynamic object removal, we introduce a robust mask generation method that enforces photometric consistency across keyframes, reducing noise from inaccurate segmentation and artifacts such as shadows. Additionally, we propose the loop-aware window selection mechanism, which utilizes unique keyframe IDs of 3D Gaussians to detect loops between the current and past frames, facilitating joint optimization of the current camera poses and the Gaussian map. DGS-SLAM achieves state-of-the-art performance in both camera tracking and novel view synthesis on various dynamic SLAM benchmarks, proving its effectiveness in handling real-world dynamic scenes.
This paper has not been read by Pith yet.
Forward citations
Cited by 5 Pith papers
-
GaussianFlow SLAM: Monocular Gaussian Splatting SLAM Guided by GaussianFlow
GaussianFlow SLAM aligns projected Gaussian motion with optical flow to regularize monocular 3D Gaussian splatting SLAM, yielding better map quality and pose accuracy than prior methods.
-
VBGS-SLAM: Variational Bayesian Gaussian Splatting Simultaneous Localization and Mapping
VBGS-SLAM uses variational inference on conjugate Gaussian properties to couple 3DGS map refinement and pose tracking with closed-form updates and posterior uncertainty, reducing drift compared to deterministic methods.
-
MoPe: Motion Permanence for Robust Monocular Gaussian Mapping in Dynamic Environments
MoPe propagates historical dynamic posteriors via SE(3) warping and bounded Bayesian fusion to maintain persistent motion state in monocular Gaussian SLAM.
-
CubifyGS: Object-Centric 3D Gaussian Splatting for Lifelong Dynamic Scene Maintenance
CubifyGS introduces object-level asset management in 3D Gaussian Splatting to handle object rearrangements in lifelong dynamic scenes more efficiently than primitive-level updates.
-
Flow4DGS-SLAM: Optical Flow-Guided 4D Gaussian Splatting SLAM
Flow4DGS-SLAM uses optical flow to generate motion masks, initialize poses, and guide 4D Gaussian modeling with scene flow and GMM for temporal properties, claiming SOTA results in dynamic tracking and reconstruction.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.