HierLoc: Hyperbolic Entity Embeddings for Hierarchical Visual Geolocation

ICLR 2026 Conference Submission9686 Authors

Keywords: Representation learning, Multimodal learning, Contrastive learning, Manifold learning, Hierarchical modeling, Geospatial AI

TL;DR: HierLoc reformulates geolocation as image-to-entity alignment in hyperbolic space, with geo-weighted contrastive learning and cross-modal attention, enabling compact inference and achieving SOTA on OSV-5M with 19.5% lower error.

Abstract: Visual geolocalization, the task of predicting where an image was taken, remains challenging due to global scale, visual ambiguity, and the inherently hierarchical structure of geography. Existing paradigms rely on either large-scale retrieval, which requires storing a large number of image embeddings, grid-based classifiers that ignore geographic continuity, or generative models that diffuse over space but struggle with fine detail. We introduce an entity-centric formulation of geolocation that replaces image-to-image retrieval with a compact hierarchy of geographic entities embedded in Hyperbolic space. Images are aligned directly to country, region, subregion, and city entities through Geo-Weighted Hyperbolic contrastive learning by directly incorporating haversine distance into the contrastive objective. This hierarchical design enables interpretable predictions and efficient inference with 240k entity embeddings instead of over 5 million image embeddings on the OSV5M benchmark, on which our method establishes a new state-of-the-art performance. Compared to the current methods in the literature, it reduces mean geodesic error by 19.5\%, while improving the fine-grained subregion accuracy by 43\%. These results demonstrate that geometry-aware hierarchical embeddings provide a scalable and conceptually new alternative for global image geolocation.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 9686