超越瓦片匹配：弥合未对齐的航空与卫星视角以实现纯视觉无人机导航
Beyond Matching to Tiles: Bridging Unaligned Aerial and Satellite Views for Vision-Only UAV Navigation
把跨视角地理定位从「匹配地图瓦片」转为「从邻域特征联合回归绝对位置与航向」，并显式编码相对空间关系以抵抗视角差异与错位，对做无人机/遥感定位与视觉导航的人是一个可直接复用的轻量化范式。
原文依据 jointly predicts UAV absolute location and heading from neighboring features
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期跨视角地理定位（CVGL）方法的进展显示出在拒止GNSS环境下支撑无人机（UAV）导航的强大潜力。然而，现有工作主要聚焦于将无人机视角与机载地图瓦片进行匹配，这带来了精度与存储开销之间的固有折衷，并且忽视了导航过程中无人机航向的重要性。此外，跨视角场景中存在的显著差异与不同程度的重叠尚未被充分考虑，限制了其在真实场景中的泛化能力。本文提出 Bearing-UAV，一种纯视觉驱动的跨视角导航方法，它从邻域特征中联合预测无人机的绝对位置与航向，从而在野外实现准确、轻量且鲁棒的导航。该方法利用全局与局部结构特征，并显式编码相对空间关系，使其对跨视角变化、错位以及特征稀疏条件具有鲁棒性。我们还提出了 Bearing-UAV-90k，一个用于评估跨视角定位与导航的多城市基准。大量实验表明，Bearing-UAV 在多种地形下取得了比以往匹配/检索范式更低的定位误差，结果令人鼓舞。我们的代码已公开于 https://github.com/liukejia121/bearinguav。
英文原文摘要 1382 字符
Recent advances in cross-view geo-localization (CVGL) methods have shown strong potential for supporting unmanned aerial vehicle (UAV) navigation in GNSS-denied environments. However, existing work predominantly focuses on matching UAV views to onboard map tiles, which introduces an inherent trade-off between accuracy and storage overhead, and overlooks the importance of the UAV's heading during navigation. Moreover, the substantial discrepancies and varying overlaps in cross-view scenarios have been insufficiently considered, limiting their generalization to real-world scenarios. In this paper, we present Bearing-UAV, a purely vision-driven cross-view navigation method that jointly predicts UAV absolute location and heading from neighboring features, enabling accurate, lightweight, and robust navigation in the wild. Our method leverages global and local structural features and explicitly encodes relative spatial relationships, making it robust to cross-view variations, misalignment, and feature-sparse conditions. We also present Bearing-UAV-90k, a multi-city benchmark for evaluating cross-view localization and navigation. Extensive experiments show encouraging results that Bearing-UAV yields lower localization errors than previous matching/retrieval paradigms across diverse terrains. Our code is publicly available at https://github.com/liukejia121/bearinguav.