Learning Cross-view Correspondences for Geo-localization on Planetary Surfaces

2026-06-29

AI Summary

This study addresses the challenge of precise geolocation on planetary surfaces where satellite-based positioning systems are unavailable, relying solely on visual matching between ground-level panoramas and overhead imagery. To this end, the authors present the first physically rendered cross-view geolocalization benchmark dataset based on a high-fidelity lunar terrain model, comprising 10,438 accurately aligned pairs of 360° ground panoramas and overhead images, augmented with overlapping tiles to enable multi-candidate, non-centered localization. Building upon this dataset, they propose and evaluate a Transformer-based cross-view retrieval method. Experimental results demonstrate that learning-based approaches can achieve effective localization in GPS-denied environments, offering a viable navigation solution for deep-space exploration independent of satellite infrastructure.

Abstract

Maintaining global position awareness is a fundamental challenge for planetary surface exploration, since satellite-based positioning systems are unavailable and onboard odometry drifts over time. Although orbital mapping products, such as overhead imagery and terrain-derived maps, provide global context, aligning them with surface observations is challenging due to large viewpoint differences, low texture, repetitive terrain, and drastic changes in appearance caused by varying illumination and topography. We introduce a new cross-view geo-localization benchmark built from physically rendered surface panoramas and overhead tiles derived from a high-resolution lunar terrain model. Our dataset contains 10438 ground views rendered as 360$^\circ$ surface panoramas with matching overhead images precisely centered at the same location. Additionally, a set of overlapping tiles is provided to study off-center localization with multiple plausible candidates per panorama. We study the performance of a state-of-the-art transformer-based geo-localization method on our data, by training it from scratch and reporting retrieval accuracy. Our results demonstrate that learning-based cross-view localization methods can be successfully applied to the domain of planetary surfaces, providing a vision-based alternative to global navigation satellite systems.

Problem

• geo-localization
• cross-view correspondence
• planetary surfaces
• orbital mapping
• viewpoint differences

Innovation

• cross-view geo-localization
• planetary surface
• physically rendered panoramas
• transformer-based localization
• lunar terrain model