TVG-SLAM: Robust Gaussian Splatting SLAM with Tri-view Geometric Constraints

2025-06-29

AI Summary

For RGB-only 3D Gaussian Splatting (3DGS) SLAM in unbounded outdoor environments, overreliance on photometric loss leads to pose drift and mapping latency. This paper proposes a robust SLAM framework grounded in triview geometric constraints. Our method jointly optimizes geometric and photometric objectives: (1) dense triview feature matching enforces geometric consistency; (2) a probabilistic Gaussian initialization strategy improves initial map quality; and (3) a render-confidence dynamic decay mechanism adaptively suppresses anomalous photometric responses. Evaluated on multiple outdoor datasets, the approach significantly enhances tracking stability and mapping accuracy—achieving a 69.0% reduction in absolute trajectory error (ATE) under the most challenging conditions—while preserving state-of-the-art rendering fidelity.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Replanning and Plan Repair

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
• Security and Privacy: Data transparency and provenance

Abstract

Recent advances in 3D Gaussian Splatting (3DGS) have enabled RGB-only SLAM systems to achieve high-fidelity scene representation. However, the heavy reliance of existing systems on photometric rendering loss for camera tracking undermines their robustness, especially in unbounded outdoor environments with severe viewpoint and illumination changes. To address these challenges, we propose TVG-SLAM, a robust RGB-only 3DGS SLAM system that leverages a novel tri-view geometry paradigm to ensure consistent tracking and high-quality mapping. We introduce a dense tri-view matching module that aggregates reliable pairwise correspondences into consistent tri-view matches, forming robust geometric constraints across frames. For tracking, we propose Hybrid Geometric Constraints, which leverage tri-view matches to construct complementary geometric cues alongside photometric loss, ensuring accurate and stable pose estimation even under drastic viewpoint shifts and lighting variations. For mapping, we propose a new probabilistic initialization strategy that encodes geometric uncertainty from tri-view correspondences into newly initialized Gaussians. Additionally, we design a Dynamic Attenuation of Rendering Trust mechanism to mitigate tracking drift caused by mapping latency. Experiments on multiple public outdoor datasets show that our TVG-SLAM outperforms prior RGB-only 3DGS-based SLAM systems. Notably, in the most challenging dataset, our method improves tracking robustness, reducing the average Absolute Trajectory Error (ATE) by 69.0% while achieving state-of-the-art rendering quality. The implementation of our method will be released as open-source.

Problem

• Enhances robustness in RGB-only SLAM under varying conditions
• Improves camera tracking with hybrid geometric constraints
• Ensures high-quality mapping via probabilistic Gaussian initialization

Innovation

• Tri-view geometry for robust SLAM tracking
• Hybrid Geometric Constraints with photometric loss
• Probabilistic Gaussian initialization using geometric uncertainty