OrthoLoC: UAV 6-DoF Localization and Calibration Using Orthographic Geodata

2025-09-22

AI Summary

To address the challenge of high-precision 6-DoF visual localization for UAVs in resource-constrained environments—characterized by GNSS denial, low-bandwidth communication, and absence of large-scale 3D models—this paper proposes a lightweight orthophoto-based cross-domain visual localization framework. Our contributions are threefold: (1) We introduce the first large-scale multimodal UAV aerial-to-orthophoto pairing dataset; (2) We design a decoupled pairing architecture enabling independent evaluation of localization and calibration performance; (3) We propose AdHoP, an adaptive optimization method integrating covisibility analysis, resolution-aware adaptation, and generic feature matching refinement. Experiments demonstrate that AdHoP achieves a 95% feature matching success rate and reduces translational error by 63%, significantly mitigating domain shift. The framework delivers robust, high-accuracy cross-domain localization and calibration across diverse real-world large-scale scenarios.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services

Abstract

Accurate visual localization from aerial views is a fundamental problem with applications in mapping, large-area inspection, and search-and-rescue operations. In many scenarios, these systems require high-precision localization while operating with limited resources (e.g., no internet connection or GNSS/GPS support), making large image databases or heavy 3D models impractical. Surprisingly, little attention has been given to leveraging orthographic geodata as an alternative paradigm, which is lightweight and increasingly available through free releases by governmental authorities (e.g., the European Union). To fill this gap, we propose OrthoLoC, the first large-scale dataset comprising 16,425 UAV images from Germany and the United States with multiple modalities. The dataset addresses domain shifts between UAV imagery and geospatial data. Its paired structure enables fair benchmarking of existing solutions by decoupling image retrieval from feature matching, allowing isolated evaluation of localization and calibration performance. Through comprehensive evaluation, we examine the impact of domain shifts, data resolutions, and covisibility on localization accuracy. Finally, we introduce a refinement technique called AdHoP, which can be integrated with any feature matcher, improving matching by up to 95% and reducing translation error by up to 63%. The dataset and code are available at: https://deepscenario.github.io/OrthoLoC.

Problem

• Accurate UAV visual localization without internet or GPS support
• Addressing domain shifts between UAV imagery and orthographic geodata
• Improving feature matching and reducing translation errors in localization

Innovation

• Uses orthographic geodata for lightweight UAV localization
• Introduces AdHoP refinement technique to improve feature matching
• Decouples image retrieval and matching for isolated performance evaluation