# GeoMetric: Injecting Metric Geographic Structure into Worldwide Image Geo-Localization

arXiv: https://arxiv.org/abs/2606.08918 · HTML: https://mlsanity.com/paper/2606.08918/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 43 references that are in the map.
- First posted on arXiv 8 Jun 2026.

## Builds on (most cited first, 17 of 17)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Google Landmarks Dataset v2 -- A Large-Scale Benchmark for Instance-Level Recognition and Retrieval](https://mlsanity.com/paper/2004.01804/) (2020)
- [Revisiting IM2GPS in the Deep Learning Era](https://mlsanity.com/paper/1705.04838/) (2017)
- [CPlaNet: Enhancing Image Geolocalization by Combinatorial Partitioning of Maps](https://mlsanity.com/paper/1808.02130/) (2018)
- [Where in the World is this Image? Transformer-based Geo-localization in the Wild](https://mlsanity.com/paper/2204.13861/) (2022)
- [Where We Are and What We're Looking At: Query Based Worldwide Image Geo-localization Using Hierarchies and Scenes](https://mlsanity.com/paper/2303.04249/) (2023)
- [G3: An Effective and Adaptive Framework for Worldwide Geolocalization Using Large Multi-Modality Models](https://mlsanity.com/paper/2405.14702/) (2024)
- [Illumination-Adaptive Person Re-identification](https://mlsanity.com/paper/1905.04525/) (2019)
- [Automatic Discovery and Geotagging of Objects from Street View Imagery](https://mlsanity.com/paper/1708.08417/) (2017)
- [MM-Path: Multi-modal, Multi-granularity Path Representation Learning -- Extended Version](https://mlsanity.com/paper/2411.18428/) (2024)
- [GRE Suite: Geo-localization Inference via Fine-Tuned Vision-Language Models and Enhanced Reasoning Chains](https://mlsanity.com/paper/2505.18700/) (2025)
- [Swarm Intelligence in Geo-Localization: A Multi-Agent Large Vision-Language Model Collaborative Framework](https://mlsanity.com/paper/2408.11312/) (2024)
- [GT-Loc: Unifying When and Where in Images Through a Joint Embedding Space](https://mlsanity.com/paper/2507.10473/) (2025)
- [GeoToken: Hierarchical Geolocalization of Images via Next Token Prediction](https://mlsanity.com/paper/2511.01082/) (2025)
- [SpotAgent: Grounding Visual Geo-localization in Large Vision-Language Models through Agentic Reasoning](https://mlsanity.com/paper/2602.09463/) (2026)
- [GeoSURGE: Geo-localization using Semantic Fusion with Hierarchy of Geographic Embeddings](https://mlsanity.com/paper/2510.01448/) (2025)

## Abstract (arXiv)

Worldwide image geo-localization aims to determine where on Earth a single image was captured. However, visually similar scenes may lie thousands of kilometers apart, so methods that localize primarily by appearance often mistake a distant look-alike for the true location. We attribute this failure to a structural cause: in existing methods, GPS coordinates serve only as training supervision, and the distance relationships among locations never enter the learned representation. To address this, we propose GeoMetric, a retrieval-based framework that encodes GPS coordinates relationally rather than in isolation, injecting the distance structure among locations into both representation learning and inference. GeoMetric comprises three components: (1) a Transformer-based GPS encoder with distance-aware location attention that modulates inter-sample aggregation by great-circle proximity; (2) a trimodal contrastive objective that aligns images, geo-textual descriptions, and GPS embeddings in a unified space; and (3) a retrieval-augmented inference stage that supplies large multimodal models (LMMs) with contrastive candidate context for grounded coordinate reasoning. Extensive experiments on IM2GPS, IM2GPS3k, YFCC4k, and YFCC26k demonstrate that GeoMetric consistently outperforms state-of-the-art methods, improving street-level accuracy (within 1 km) by 1.5%, 0.9%, 6.9%, and 2.5%, respectively. Controlled ablations confirm that the gains originate from the proposed geographic encoding rather than any specific LMM.