GeoExplorer: Active Geo-localization with Curiosity-Driven Exploration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mi, Li, Bechaz, Manon, Chen, Zeming, Bosselut, Antoine, Tuia, Devis
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916874921443328
author Mi, Li
Bechaz, Manon
Chen, Zeming
Bosselut, Antoine
Tuia, Devis
author_facet Mi, Li
Bechaz, Manon
Chen, Zeming
Bosselut, Antoine
Tuia, Devis
contents Active Geo-localization (AGL) is the task of localizing a goal, represented in various modalities (e.g., aerial images, ground-level images, or text), within a predefined search area. Current methods approach AGL as a goal-reaching reinforcement learning (RL) problem with a distance-based reward. They localize the goal by implicitly learning to minimize the relative distance from it. However, when distance estimation becomes challenging or when encountering unseen targets and environments, the agent exhibits reduced robustness and generalization ability due to the less reliable exploration strategy learned during training. In this paper, we propose GeoExplorer, an AGL agent that incorporates curiosity-driven exploration through intrinsic rewards. Unlike distance-based rewards, our curiosity-driven reward is goal-agnostic, enabling robust, diverse, and contextually relevant exploration based on effective environment modeling. These capabilities have been proven through extensive experiments across four AGL benchmarks, demonstrating the effectiveness and generalization ability of GeoExplorer in diverse settings, particularly in localizing unfamiliar targets and environments.
format Preprint
id arxiv_https___arxiv_org_abs_2508_00152
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GeoExplorer: Active Geo-localization with Curiosity-Driven Exploration
Mi, Li
Bechaz, Manon
Chen, Zeming
Bosselut, Antoine
Tuia, Devis
Computer Vision and Pattern Recognition
Active Geo-localization (AGL) is the task of localizing a goal, represented in various modalities (e.g., aerial images, ground-level images, or text), within a predefined search area. Current methods approach AGL as a goal-reaching reinforcement learning (RL) problem with a distance-based reward. They localize the goal by implicitly learning to minimize the relative distance from it. However, when distance estimation becomes challenging or when encountering unseen targets and environments, the agent exhibits reduced robustness and generalization ability due to the less reliable exploration strategy learned during training. In this paper, we propose GeoExplorer, an AGL agent that incorporates curiosity-driven exploration through intrinsic rewards. Unlike distance-based rewards, our curiosity-driven reward is goal-agnostic, enabling robust, diverse, and contextually relevant exploration based on effective environment modeling. These capabilities have been proven through extensive experiments across four AGL benchmarks, demonstrating the effectiveness and generalization ability of GeoExplorer in diverse settings, particularly in localizing unfamiliar targets and environments.
title GeoExplorer: Active Geo-localization with Curiosity-Driven Exploration
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.00152