Search papers, labs, and topics across Lattice.
This paper introduces OpenCVL, a large-scale dataset designed for Fine-Grained Cross-View Localization (CVL), which consists of 617,388 ground-aerial image pairs from 41 cities across four European countries. The dataset addresses the limitations of existing high-end sensor-based datasets by incorporating diverse in-the-wild imagery, while a robust data curation framework ensures reliable pose annotations. Experiments reveal that integrating noisy in-the-wild data enhances the performance of state-of-the-art CVL models on clean test sets, highlighting the potential for improved localization in urban environments.
Incorporating noisy, in-the-wild imagery into CVL models significantly boosts their performance on clean datasets, challenging traditional reliance on high-end sensors.
Fine-grained Cross-View Localization (CVL) estimates the precise position and orientation of a ground-level image by aligning it with geo-referenced aerial imagery, offering a scalable alternative to Global Navigation Satellite Systems (GNSS) in challenging urban environments. Existing datasets rely on data collected with high-end sensor suites, which inherently limit image diversity and scalability. While in-the-wild images are abundant, their noisy geo-tags make them unsuitable for reliable evaluation. To bridge this gap, we introduce OpenCVL, a large-scale, diverse, and open dataset containing 617,388 ground-aerial image pairs spanning 41 cities across four European countries. All images are sourced from permissive platforms, ensuring long-term accessibility and supporting open and reproducible research. The training set combines images captured with high-end sensors with diverse in-the-wild imagery. We further develop a data curation framework that filters and corrects pose annotations to construct reliable in-the-wild evaluation data. In addition, OpenCVL includes dedicated cross-area and snowy test sets to assess generalization and robustness. Experiments with a state-of-the-art CVL model on OpenCVL show that incorporating noisy in-the-wild data consistently improves performance on clean test sets, suggesting a promising direction for scaling CVL with diverse real-world imagery.
