VICI: VLM-Instructed Cross-view Image-localisation

TL;DR

This paper introduces a two-stage retrieval and re-ranking method for cross-view image localisation that improves matching accuracy between limited-FOV street images and satellite images, addressing practical real-world scenarios.

Contribution

It proposes a novel two-stage retrieval and re-ranking approach tailored for limited-FOV images, enhancing practical geo-localisation performance.

Findings

Achieved R@1 of approximately 45% and R@10 of approximately 75%.

Demonstrated the effectiveness of re-ranking in improving retrieval accuracy.

Addressed the challenge of matching limited-FOV images with satellite imagery in real-world conditions.

Abstract

In this paper, we present a high-performing solution to the UAVM 2025 Challenge, which focuses on matching narrow FOV street-level images to corresponding satellite imagery using the University-1652 dataset. As panoramic Cross-View Geo-Localisation nears peak performance, it becomes increasingly important to explore more practical problem formulations. Real-world scenarios rarely offer panoramic street-level queries; instead, queries typically consist of limited-FOV images captured with unknown camera parameters. Our work prioritises discovering the highest achievable performance under these constraints, pushing the limits of existing architectures. Our method begins by retrieving candidate satellite image embeddings for a given query, followed by a re-ranking stage that selectively enhances retrieval accuracy within the top candidates. This two-stage approach enables more precise matching, even under the significant viewpoint and scale variations inherent in the task. Through experimentation, we demonstrate that our approach achieves competitive results -specifically attaining R@1 and R@10 retrieval rates of \topone\% and \topten\% respectively. This underscores the potential of optimised retrieval and re-ranking strategies in advancing practical geo-localisation performance. Code is available at https://github.com/tavisshore/VICI.