Reproduce “Warp-free Cross-view Geo-localization via Feature-space Consensus Mining” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhuo Song, Lian Xu, Runqing Jiang, Yongjian Zhang, Kunhong Li, Ye Zhang, Yulan Guo
GeoCoM learns cross-view agreement for matching ground-level or UAV images to geo-referenced satellite images. During training, paired views form a joint representation, and shared pattern probes project each view and the joint representation into a common feature space for contrastive alignment. On VIGOR cross-area retrieval, GeoCoM exceeds Co-Retrieval at R@1 and R@5, but not R@1%.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.