Metraipolis › archive › dc_a2a63450eb8d
Anchored Rank-Adversarial Fine-Tuning Pretrained Visual Place Recognition Models — arXiv:2611.05286
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.05286
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T19:04:57.533Z
- SHA-256 of the archived text
b8b62298da1ac9b6d1aefbaabad69e5d78e7ee63d82de852620d14bc3696134a- Stable URL of this page
https://metraipolis.com/documento/dc_a2a63450eb8d
How to cite
Anchored Rank-Adversarial Fine-Tuning Pretrained Visual Place Recognition Models — arXiv:2611.05286 — arXiv (en), https://arxiv.org/abs/2611.05286 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a2a63450eb8d (Metraipolis archive dc_a2a63450eb8d, sha256 b8b62298da1ac9b6d1aefbaabad69e5d78e7ee63d82de852620d14bc3696134a)
BibTeX
@misc{dca2a63450eb8d,
title = {Anchored Rank-Adversarial Fine-Tuning Pretrained Visual Place Recognition Models — arXiv:2611.05286},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.05286},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a2a63450eb8d; sha256 b8b62298da1ac9b6d1aefbaabad69e5d78e7ee63d82de852620d14bc3696134a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05286. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Visual place recognition (VPR) localises a camera by retrieving similar geo-tagged images, and current foundation-model VPR models are accurate across cities, seasons and illumination, but not robust to adversarial perturbations: under a white-box attack on the ranking at an imperceptible budget of 1/255, SuperVLAD and BoQ keep at most 0.2% top-1 recall (R@1) on four benchmarks. We propose anchored rank-adversarial fine-tuning (ARAFT), which fine-tunes a released model with a listwise rank loss on each adversarial query's hardest positive and an anchor to the clean descriptor of the frozen pretrained model. Checkpoints are selected under a clean-accuracy budget fixed in advance, and the full trade-off is reported. Within five validation R@1 points of clean cost, plain adversarial training has no admissible checkpoint, whereas ARAFT raises robust R@1 from zero to 47.3 on SuperVLAD and 55.3 on BoQ; plain adversarial training exceeds this only at about three times the clean cost. Against unbudgeted plain adversarial training on the test sets, ARAFT matches its robust R@1 at 2/255 with about ten more clean R@1 points. At 4/255, plain adversarial training is 6-9 robust R@1 points ahead, while ARAFT still keeps its ten-point clean advantage. Developed on SuperVLAD, ARAFT transfers to BoQ without re-tuning, and its robustness holds against stronger optimisers, with no sign of gradient masking. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2611.05286 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-04T18:39:37.585Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).