Image-based Geo-localization for Robotics: Are Black-box Vision-Language Models there yet?

TL;DR

This paper systematically investigates the potential of state-of-the-art black-box Vision-Language Models for zero-shot image geo-localization, considering realistic constraints and different query scenarios, providing new insights into their capabilities.

Contribution

It is the first comprehensive study exploring the use of black-box VLMs as standalone, zero-shot geo-localization systems with various prompt and query scenarios.

Findings

VLMs show potential in zero-shot geo-localization tasks.

Semantic equivalence in prompts can improve localization accuracy.

Model consistency is a useful metric alongside traditional accuracy.

Abstract

The advances in Vision-Language models (VLMs) offer exciting opportunities for robotic applications involving image geo-localization, the problem of identifying the geo-coordinates of a place based on visual data only. Recent research works have focused on using a VLM as embeddings extractor for geo-localization, however, the most sophisticated VLMs may only be available as black boxes that are accessible through an API, and come with a number of limitations: there is no access to training data, model features and gradients; retraining is not possible; the number of predictions may be limited by the API; training on model outputs is often prohibited; and queries are open-ended. The utilization of a VLM as a stand-alone, zero-shot geo-localization system using a single text-based prompt is largely unexplored. To bridge this gap, this paper undertakes the first systematic study, to the best of our knowledge, to investigate the potential of some of the state-of-the-art VLMs as stand-alone, zero-shot geo-localization systems in a black-box setting with realistic constraints. We consider three main scenarios for this thorough investigation: a) fixed text-based prompt; b) semantically-equivalent text-based prompts; and c) semantically-equivalent query images. We also take into account the auto-regressive and probabilistic generation process of the VLMs when investigating their utility for geo-localization task by using model consistency as a metric in addition to traditional accuracy. Our work provides new insights in the capabilities of different VLMs for the above-mentioned scenarios.