Where Do Vision-Language Models Fail? World Scale Analysis for Image Geolocalization

TL;DR

This paper systematically evaluates state-of-the-art Vision-Language Models for zero-shot country-level image geolocalization, revealing their strengths in semantic reasoning and limitations in fine-grained geographic cues.

Contribution

It provides the first focused comparison of VLMs for country-level geolocalization, highlighting their potential and current limitations without task-specific training.

Findings

Significant variation in model performance across datasets

Semantic reasoning aids coarse geolocalization

Current VLMs struggle with fine-grained geographic cues

Abstract

Image geolocalization has traditionally been addressed through retrieval-based place recognition or geometry-based visual localization pipelines. Recent advances in Vision-Language Models (VLMs) have demonstrated strong zero-shot reasoning capabilities across multimodal tasks, yet their performance in geographic inference remains underexplored. In this work, we present a systematic evaluation of multiple state-of-the-art VLMs for country-level image geolocalization using ground-view imagery only. Instead of relying on image matching, GPS metadata, or task-specific training, we evaluate prompt-based country prediction in a zero-shot setting. The selected models are tested on three geographically diverse datasets to assess their robustness and generalization ability. Our results reveal substantial variation across models, highlighting the potential of semantic reasoning for coarse geolocalization and the limitations of current VLMs in capturing fine-grained geographic cues. This study provides the first focused comparison of modern VLMs for country-level geolocalization and establishes a foundation for future research at the intersection of multimodal reasoning and geographic understanding.