Cross-view geo-localization: a survey

TL;DR

This survey reviews recent advances in cross-view geo-localization, highlighting feature-based and deep learning methods, challenges like viewpoint variations, and benchmarking datasets, to guide future research in this evolving field.

Contribution

It provides a comprehensive overview of methodologies, challenges, datasets, and evaluation metrics, offering insights and future directions in cross-view geo-localization research.

Findings

Deep learning methods outperform feature-based approaches in accuracy.

Viewpoint and illumination variations remain significant challenges.

Benchmark datasets enable standardized evaluation of techniques.

Abstract

Cross-view geo-localization has garnered notable attention in the realm of computer vision, spurred by the widespread availability of copious geotagged datasets and the advancements in machine learning techniques. This paper provides a thorough survey of cutting-edge methodologies, techniques, and associated challenges that are integral to this domain, with a focus on feature-based and deep learning strategies. Feature-based methods capitalize on unique features to establish correspondences across disparate viewpoints, whereas deep learning-based methodologies deploy convolutional neural networks to embed view-invariant attributes. This work also delineates the multifaceted challenges encountered in cross-view geo-localization, such as variations in viewpoints and illumination, the occurrence of occlusions, and it elucidates innovative solutions that have been formulated to tackle these issues. Furthermore, we delineate benchmark datasets and relevant evaluation metrics, and also perform a comparative analysis of state-of-the-art techniques. Finally, we conclude the paper with a discussion on prospective avenues for future research and the burgeoning applications of cross-view geo-localization in an intricately interconnected global landscape.