LSGS-Loc: Towards Robust 3DGS-Based Visual Localization for Large-Scale UAV Scenarios

2026-04-06

AI Summary

This work addresses the challenges of visual localization in large-scale drone scenes using 3D Gaussian Splatting (3DGS), where performance is often compromised by unstable pose initialization and rendering artifacts. To overcome these issues, the authors propose LSGS-Loc, a robust, scene-agnostic localization pipeline that requires no scene-specific training. LSGS-Loc achieves scale-aware pose initialization by integrating scene-independent relative pose estimation with explicit scale constraints derived from 3DGS geometry. Furthermore, it introduces a Laplacian reliability mask to suppress reconstruction artifacts—such as blur and floating elements—during photometric optimization. Evaluated on large-scale drone benchmarks, LSGS-Loc significantly outperforms existing 3DGS-based localization methods, achieving state-of-the-art accuracy and robustness on unordered image queries.

Abstract

Visual localization in large-scale UAV scenarios is a critical capability for autonomous systems, yet it remains challenging due to geometric complexity and environmental variations. While 3D Gaussian Splatting (3DGS) has emerged as a promising scene representation, existing 3DGS-based visual localization methods struggle with robust pose initialization and sensitivity to rendering artifacts in large-scale settings. To address these limitations, we propose LSGS-Loc, a novel visual localization pipeline tailored for large-scale 3DGS scenes. Specifically, we introduce a scale-aware pose initialization strategy that combines scene-agnostic relative pose estimation with explicit 3DGS scale constraints, enabling geometrically grounded localization without scene-specific training. Furthermore, in the pose refinement, to mitigate the impact of reconstruction artifacts such as blur and floaters, we develop a Laplacian-based reliability masking mechanism that guides photometric refinement toward high-quality regions. Extensive experiments on large-scale UAV benchmarks demonstrate that our method achieves state-of-the-art accuracy and robustness for unordered image queries, significantly outperforming existing 3DGS-based approaches. Code is available at: https://github.com/xzhang-z/LSGS-Loc

Problem

• visual localization
• 3D Gaussian Splatting
• UAV
• pose initialization
• rendering artifacts

Innovation

• 3D Gaussian Splatting
• visual localization
• pose initialization
• reliability masking
• UAV