GeoVLM: Improving Automated Vehicle Geolocalisation Using Vision-Language Matching

TL;DR

GeoVLM leverages vision-language models and interpretable language descriptions to enhance cross-view geo-localisation accuracy, addressing scene similarity challenges and improving top match retrieval in automated vehicle positioning.

Contribution

Introduces GeoVLM, a novel trainable reranking method using zero-shot vision-language models and natural language descriptions for improved geo-localisation accuracy.

Findings

GeoVLM outperforms state-of-the-art methods on benchmark datasets.

It improves top match retrieval accuracy significantly.

The approach is effective in real-world driving environments.

Abstract

Cross-view geo-localisation identifies coarse geographical position of an automated vehicle by matching a ground-level image to a geo-tagged satellite image from a database. Despite the advancements in Cross-view geo-localisation, significant challenges still persist such as similar looking scenes which makes it challenging to find the correct match as the top match. Existing approaches reach high recall rates but they still fail to rank the correct image as the top match. To address this challenge, this paper proposes GeoVLM, a novel approach which uses the zero-shot capabilities of vision language models to enable cross-view geo-localisation using interpretable cross-view language descriptions. GeoVLM is a trainable reranking approach which improves the best match accuracy of cross-view geo-localisation. GeoVLM is evaluated on standard benchmark VIGOR and University-1652 and also through real-life driving environments using Cross-View United Kingdom, a new benchmark dataset introduced in this paper. The results of the paper show that GeoVLM improves retrieval performance of cross-view geo-localisation compared to the state-of-the-art methods with the help of explainable natural language descriptions. The code is available at https://github.com/CAV-Research-Lab/GeoVLM