Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.02737 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911297988198400 |
|---|---|
| author | Amadei, Tristan Meinhardt-Llopis, Enric Bascle, Benedicte Abgrall, Corentin Facciolo, Gabriele |
| author_facet | Amadei, Tristan Meinhardt-Llopis, Enric Bascle, Benedicte Abgrall, Corentin Facciolo, Gabriele |
| contents | Image-based localization in GNSS-denied environments is critical for UAV autonomy. Existing state-of-the-art approaches rely on matching UAV images to geo-referenced satellite images; however, they typically require large-scale, paired UAV-satellite datasets for training. Such data are costly to acquire and often unavailable, limiting their applicability. To address this challenge, we adopt a training paradigm that removes the need for UAV imagery during training by learning directly from satellite-view reference images. This is achieved through a dedicated augmentation strategy that simulates the visual domain shift between satellite and real-world UAV views. We introduce CAEVL, an efficient model designed to exploit this paradigm, and validate it on ViLD, a new and challenging dataset of real-world UAV images that we release to the community. Our method achieves competitive performance compared to approaches trained with paired data, demonstrating its effectiveness and strong generalization capabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_02737 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Paired Data: Self-Supervised UAV Geo-Localization from Reference Imagery Alone Amadei, Tristan Meinhardt-Llopis, Enric Bascle, Benedicte Abgrall, Corentin Facciolo, Gabriele Computer Vision and Pattern Recognition Machine Learning Image-based localization in GNSS-denied environments is critical for UAV autonomy. Existing state-of-the-art approaches rely on matching UAV images to geo-referenced satellite images; however, they typically require large-scale, paired UAV-satellite datasets for training. Such data are costly to acquire and often unavailable, limiting their applicability. To address this challenge, we adopt a training paradigm that removes the need for UAV imagery during training by learning directly from satellite-view reference images. This is achieved through a dedicated augmentation strategy that simulates the visual domain shift between satellite and real-world UAV views. We introduce CAEVL, an efficient model designed to exploit this paradigm, and validate it on ViLD, a new and challenging dataset of real-world UAV images that we release to the community. Our method achieves competitive performance compared to approaches trained with paired data, demonstrating its effectiveness and strong generalization capabilities. |
| title | Beyond Paired Data: Self-Supervised UAV Geo-Localization from Reference Imagery Alone |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2512.02737 |