ViewBridge:Revisiting Cross-View Localization from Image Matching
This work addresses cross-view localization for applications like autonomous navigation and mapping, presenting an incremental improvement through novel components for geometric consistency and match refinement.
The paper tackles the problem of cross-view localization (estimating ground image poses from aerial imagery) by addressing limitations in existing methods that fail to establish reliable correspondences under large viewpoint variations. The proposed framework achieves geometry-consistent and fine-grained correspondences, improving localization accuracy and stability.
Cross-view localization aims to estimate the 3-DoF pose of a ground-view image by aligning it with aerial or satellite imagery. Existing methods typically address this task through direct regression or feature alignment in a shared bird's-eye view (BEV) space. Although effective for coarse alignment, these methods fail to establish fine-grained and geometrically reliable correspondences under large viewpoint variations, thereby limiting both the accuracy and interpretability of localization results. Consequently, we revisit cross-view localization from the perspective of image matching and propose a unified framework that enhances both matching and localization. Specifically, we introduce a Surface Model that constrains BEV feature projection to physically valid regions for geometric consistency, and a SimRefiner that adaptively refines similarity distributions to enhance match reliability. To further support research in this area, we present CVFM, the first benchmark with 32,509 cross-view image pairs annotated with pixel-level correspondences. Extensive experiments demonstrate that our approach achieves geometry-consistent and fine-grained correspondences across extreme viewpoints and further improves the accuracy and stability of cross-view localization.