D$^2$GSLAM: 4D Dynamic Gaussian Splatting SLAM

2025-12-10

AI Summary

To address the challenge in dynamic environments where dense SLAM struggles to simultaneously achieve high-accuracy static mapping and faithful modeling of dynamic object motion, this paper proposes the first 4D Gaussian SLAM framework enabling joint high-fidelity reconstruction of static scenes and dynamic objects, along with robust camera tracking. Methodologically, it introduces a novel integration of geometry-guided dynamic segmentation, co-optimized static-dynamic 4D Gaussian representations, progressive pose refinement, and motion-consistency loss, further enhanced by multi-view geometric constraints and temporal motion modeling for joint optimization. Compared to prior approaches, our method preserves physically plausible trajectories of dynamic objects while significantly improving both camera tracking accuracy and scene reconstruction fidelity, achieving state-of-the-art performance across multiple dynamic SLAM benchmarks.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Recent advances in Dense Simultaneous Localization and Mapping (SLAM) have demonstrated remarkable performance in static environments. However, dense SLAM in dynamic environments remains challenging. Most methods directly remove dynamic objects and focus solely on static scene reconstruction, which ignores the motion information contained in these dynamic objects. In this paper, we present D$^2$GSLAM, a novel dynamic SLAM system utilizing Gaussian representation, which simultaneously performs accurate dynamic reconstruction and robust tracking within dynamic environments. Our system is composed of four key components: (i) We propose a geometric-prompt dynamic separation method to distinguish between static and dynamic elements of the scene. This approach leverages the geometric consistency of Gaussian representation and scene geometry to obtain coarse dynamic regions. The regions then serve as prompts to guide the refinement of the coarse mask for achieving accurate motion mask. (ii) To facilitate accurate and efficient mapping of the dynamic scene, we introduce dynamic-static composite representation that integrates static 3D Gaussians with dynamic 4D Gaussians. This representation allows for modeling the transitions between static and dynamic states of objects in the scene for composite mapping and optimization. (iii) We employ a progressive pose refinement strategy that leverages both the multi-view consistency of static scene geometry and motion information from dynamic objects to achieve accurate camera tracking. (iv) We introduce a motion consistency loss, which leverages the temporal continuity in object motions for accurate dynamic modeling. Our D$^2$GSLAM demonstrates superior performance on dynamic scenes in terms of mapping and tracking accuracy, while also showing capability in accurate dynamic modeling.

Problem

• Simultaneously reconstructs dynamic objects and tracks camera in dynamic environments
• Separates static and dynamic scene elements using geometric-prompt dynamic separation
• Models static and dynamic states with composite Gaussian representation for mapping

Innovation

• Geometric-prompt dynamic separation method for motion mask
• Dynamic-static composite Gaussian representation for mapping
• Progressive pose refinement using static and dynamic cues