A Novel Geo-Localization Method for UAV and Satellite Images Using Cross-View Consistent Attention

Geo-localization has been widely applied as an important technique to get the longitudeand latitude for unmanned aerial vehicle (UAV) navigation in outdoor flight. Due to the possibleinterference and blocking of GPS signals, the method based on image retrieval, which is less likelyto be interfered with, has received extensive attention in recent years. The geo-localization of UAVsand satellites can be achieved by querying pre-obtained satellite images with GPS-tagged and droneimages from different perspectives. In this paper, an image transformation technique is used to extractcross-view geo-localization information from UAVs and satellites. A single-stage training methodin UAV and satellite geo-localization is first proposed, which simultaneously realizes cross-viewfeature extraction and image retrieval, and achieves higher accuracy than existing multi-stage trainingtechniques. A novel piecewise soft-margin triplet loss function is designed to avoid model parametersbeing trapped in suboptimal sets caused by the lack of constraint on positive and negative samples.The results illustrate that the proposed loss function enhances image retrieval accuracy and realizesa better convergence. Moreover, a data augmentation method for satellite images is proposed toovercome the disproportionate numbers of image samples. On the benchmark University-1652, theproposed method achieves the state-of-the-art result with a 6.67% improvement in recall rate (R@1)and 6.13% in average precision (AP). All codes will be publicized to promote reproducibility.