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.

Visual place recognition (VPR) is an essential component of robot navigation and localization systems that allows them to identify a place using only image data. VPR is challenging due to the significant changes in a place's appearance under different illumination throughout the day, with seasonal weather and when observed from different viewpoints. Currently, no single VPR technique excels in every environmental condition, each exhibiting unique benefits and shortcomings. As a result, VPR systems combining multiple techniques achieve more reliable VPR performance in changing environments, at the cost of higher computational loads. Addressing this shortcoming, we propose an adaptive VPR system dubbed Adaptive Multi-Self Identification and Correction (A-MuSIC). We start by developing a method to collect information of the runtime performance of a VPR technique by analysing the frame-to-frame continuity of matched queries. We then demonstrate how to operate the method on a static ensemble of techniques, generating data on which techniques are contributing the most for the current environment. A-MuSIC uses the collected information to both select a minimal subset of techniques and to decide when a re-selection is required during navigation. A-MuSIC matches or beats state-of-the-art VPR performance across all tested benchmark datasets while maintaining its computational load on par with individual techniques.