CDIKTNet uses limited paired data to initialize cross-view correlations via a cross-domain invariance sub-network and refines them with dual-path contrastive learning in a transfer sub-network, claiming SOTA results on drone-view geo-localization under full, few-shot, and cross-domain settings.
University-1652: A multi-view multi- source benchmark for drone-based geo-localization,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From Limited Labels to Open Domains:An Efficient Learning Method for Drone-view Geo-Localization
CDIKTNet uses limited paired data to initialize cross-view correlations via a cross-domain invariance sub-network and refines them with dual-path contrastive learning in a transfer sub-network, claiming SOTA results on drone-view geo-localization under full, few-shot, and cross-domain settings.