Ground-to-Satellite Localization in Unconstrained Image Collections for 3D Scene Reconstruction

arXiv:2608.29211v1 Announce Type: new Abstract: Ground image localization with respect to satellite imagery is a key enabler for metrically-accurate, geo-localized 3D scene reconstruction from unconstrained image collections. Existing cross-view localization methods have strict requirements such as panoramic imagery or known initial locations, limiting their applicability for in-the-wild reconstruction settings. We propose a robust hierarchical cross-view localization framework that leverages geometric constraints from Structure-from-Motion (SfM) models derived from unconstrained ground image collections. Our method generates coarse-to-fine pose hypotheses through a cross-view matching approach and aggregates noisy predictions across SfM model(s) using Kernel Density Estimation to recover consensus alignments while filtering outliers. Experiments demonstrate reliable localization performance from challenging image collections. Empirically we found satellite-referenced alignment enables accurate metric scale estimation, doppelganger detection, and merging of disjoint SfM reconstructions, resulting in more complete, geo-localized site models than are possible with SfM alone.

Related

• Seeing Across Skies and Streets: Feedforward 3D Reconstruction from Satellite, Drone, and Ground Images
• OpenCVL: An Open, Diverse, and Large-Scale Dataset for Fine-Grained Cross-View Localization
• Recurrent Cross-View Object Geo-Localization
• TACO: Trajectory Aligning Cross-view Optimisation

Source: arXiv cs.CV | 2026-09-01