{"ID":6497851,"CreatedAt":"2026-07-13T01:19:40.13847098Z","UpdatedAt":"2026-07-14T01:36:59.12045529Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.09057","arxiv_id":"2607.09057","title":"STEAM: Stable Self-Training with Elastic Matching and Adaptive Purification","abstract":"Cross-view geo-localization (CVGL) aims to achieve GPS-free localization by matching drone-view images with corresponding satellite-view images. Existing supervised methods rely on large-scale manually annotated cross-view image pairs, making them costly and difficult to scale. In contrast, existing unsupervised approaches typically depend on generative models or clustering-based stage-wise optimization, which are prone to distribution bias and the accumulation of noisy pseudo-labels. To address these limitations, we propose STEAM (Stable Self-Training with Elastic Matching and Adaptive Purification), an end-to-end unsupervised cross-view geo-localization framework that performs self-training directly on real drone and satellite images. Specifically, the proposed Stable Spatial-Aware Module enhances the stability of feature representations, Elastic Matching discovers high-quality cross-view pseudo-labels, and Adaptive Purification dynamically maintains a reliable pseudo-label repository throughout the self-training process. Extensive experiments on the University-1652 and SUES-200 benchmarks demonstrate that STEAM achieves state-of-the-art performance among all existing unsupervised methods and delivers performance comparable to supervised approaches, validating the effectiveness and superiority of the proposed framework. The source code is available at https://github.com/wsx-heu/STEAM.git.","short_abstract":"Cross-view geo-localization (CVGL) aims to achieve GPS-free localization by matching drone-view images with corresponding satellite-view images. Existing supervised methods rely on large-scale manually annotated cross-view image pairs, making them costly and difficult to scale. In contrast, existing unsupervised approa...","url_abs":"https://arxiv.org/abs/2607.09057","url_pdf":"https://arxiv.org/pdf/2607.09057v1","authors":"[\"Shaoxiang Wang\",\"Kejia Zhang\",\"Haiwei Pan\",\"Lan Zhang\"]","published":"2026-07-10T02:56:34Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":614130,"CreatedAt":"2026-07-13T01:19:40.13847098Z","UpdatedAt":"2026-07-13T01:19:40.13847098Z","DeletedAt":null,"paper_id":6497851,"paper_url":"https://arxiv.org/abs/2607.09057","paper_title":"STEAM: Stable Self-Training with Elastic Matching and Adaptive Purification","repo_url":"https://github.com/wsx-heu/STEAM.git","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}