DualGeo: A Dual-View Framework for Worldwide Image Geo-localization

TL;DR

DualGeo is a two-stage worldwide image geo-localization framework that combines feature fusion, contrastive learning, and large multimodal models to improve localization accuracy across various scales.

Contribution

It introduces a novel dual-view framework that fuses image and semantic features, and employs geo-cognitive refinement with multimodal models for enhanced accuracy.

Findings

Outperforms state-of-the-art methods on IM2GPS, IM2GPS3k, and YFCC4k datasets.

Improves street-level localization accuracy by up to 16.58%.

Enhances city-level localization accuracy by up to 8.77%.

Abstract

Worldwide image geo-localization aims to infer the geographic location of an image captured anywhere on Earth, spanning street, city, regional, national, and continental scales. Existing methods rely on visual features that are sensitive to environmental variations (e.g., lighting, season, and weather) and lack effective post-processing to filter outlier candidates, limiting localization accuracy. To address these limitations, we propose DualGeo, a two-stage framework for worldwide image geo-localization. First, it establishes a geo-representational foundation by fusing image and semantic segmentation features via bidirectional cross-attention. The fused features are then aligned with GPS coordinates through dual-view contrastive learning to build a global retrieval database. Second, it performs geo-cognitive refinement by re-ranking retrieved candidates using geographic clustering. It then feeds them into large multimodal models (LMMs) for final coordinate prediction. Experiments on IM2GPS, IM2GPS3k, and YFCC4k show that DualGeo outperforms state-of-the-art methods, improving street-level (<1 km) and city-level (<25 km) localization accuracy by 3.6%-16.58% and 1.29%-8.77%, respectively. Our code and datasets are available : https://github.com/CJ310177/DualGeo.