Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Xiaowei, Wang, Di, Li, Ke, Wang, Yifeng, Wang, Chengjian, Sun, Libin, Wu, Zhihong, Zhang, Yiming, Wang, Quan
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2509.20684
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914055801798656
author Wang, Xiaowei
Wang, Di
Li, Ke
Wang, Yifeng
Wang, Chengjian
Sun, Libin
Wu, Zhihong
Zhang, Yiming
Wang, Quan
author_facet Wang, Xiaowei
Wang, Di
Li, Ke
Wang, Yifeng
Wang, Chengjian
Sun, Libin
Wu, Zhihong
Zhang, Yiming
Wang, Quan
contents Cross-view geo-localization (CVGL) aims to match images of the same location captured from drastically different viewpoints. Despite recent progress, existing methods still face two key challenges: (1) achieving robustness under severe appearance variations induced by diverse UAV orientations and fields of view, which hinders cross-domain generalization, and (2) establishing reliable correspondences that capture both global scene-level semantics and fine-grained local details. In this paper, we propose EGS, a novel CVGL framework designed to enhance cross-domain generalization. Specifically, we introduce an E(2)-Steerable CNN encoder to extract stable and reliable features under rotation and viewpoint shifts. Furthermore, we construct a graph with a virtual super-node that connects to all local nodes, enabling global semantics to be aggregated and redistributed to local regions, thereby enforcing global-local consistency. Extensive experiments on the University-1652 and SUES-200 benchmarks demonstrate that EGS consistently achieves substantial performance gains and establishes a new state of the art in cross-domain CVGL.
format Preprint
id arxiv_https___arxiv_org_abs_2509_20684
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhancing Cross-View Geo-Localization Generalization via Global-Local Consistency and Geometric Equivariance
Wang, Xiaowei
Wang, Di
Li, Ke
Wang, Yifeng
Wang, Chengjian
Sun, Libin
Wu, Zhihong
Zhang, Yiming
Wang, Quan
Computer Vision and Pattern Recognition
Cross-view geo-localization (CVGL) aims to match images of the same location captured from drastically different viewpoints. Despite recent progress, existing methods still face two key challenges: (1) achieving robustness under severe appearance variations induced by diverse UAV orientations and fields of view, which hinders cross-domain generalization, and (2) establishing reliable correspondences that capture both global scene-level semantics and fine-grained local details. In this paper, we propose EGS, a novel CVGL framework designed to enhance cross-domain generalization. Specifically, we introduce an E(2)-Steerable CNN encoder to extract stable and reliable features under rotation and viewpoint shifts. Furthermore, we construct a graph with a virtual super-node that connects to all local nodes, enabling global semantics to be aggregated and redistributed to local regions, thereby enforcing global-local consistency. Extensive experiments on the University-1652 and SUES-200 benchmarks demonstrate that EGS consistently achieves substantial performance gains and establishes a new state of the art in cross-domain CVGL.
title Enhancing Cross-View Geo-Localization Generalization via Global-Local Consistency and Geometric Equivariance
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.20684