Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.14674 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912667209302016 |
|---|---|
| author | Li, Ling Zhou, Yao Liang, Yuxuan Tsung, Fugee Wei, Jiaheng |
| author_facet | Li, Ling Zhou, Yao Liang, Yuxuan Tsung, Fugee Wei, Jiaheng |
| contents | Previous methods for image geo-localization have typically treated the task as either classification or retrieval, often relying on black-box decisions that lack interpretability. The rise of large vision-language models (LVLMs) has enabled a rethinking of geo-localization as a reasoning-driven task grounded in visual cues. However, two major challenges persist. On the data side, existing reasoning-focused datasets are primarily based on street-view imagery, offering limited scene diversity and constrained viewpoints. On the modeling side, current approaches predominantly rely on supervised fine-tuning, which yields only marginal improvements in reasoning capabilities. To address these challenges, we propose a novel pipeline that constructs a reasoning-oriented geo-localization dataset, MP16-Reason, using diverse social media images. We introduce GLOBE, Group-relative policy optimization for Localizability assessment and Optimized visual-cue reasoning, yielding Bi-objective geo-Enhancement for the VLM in recognition and reasoning. GLOBE incorporates task-specific rewards that jointly enhance localizability assessment, visual-cue reasoning, and geolocation accuracy. Both qualitative and quantitative results demonstrate that GLOBE outperforms state-of-the-art open-source LVLMs on geo-localization tasks, particularly in diverse visual scenes, while also generating more insightful and interpretable reasoning trajectories. The data and code are available at https://github.com/lingli1996/GLOBE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_14674 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Recognition through Reasoning: Reinforcing Image Geo-localization with Large Vision-Language Models Li, Ling Zhou, Yao Liang, Yuxuan Tsung, Fugee Wei, Jiaheng Computer Vision and Pattern Recognition Previous methods for image geo-localization have typically treated the task as either classification or retrieval, often relying on black-box decisions that lack interpretability. The rise of large vision-language models (LVLMs) has enabled a rethinking of geo-localization as a reasoning-driven task grounded in visual cues. However, two major challenges persist. On the data side, existing reasoning-focused datasets are primarily based on street-view imagery, offering limited scene diversity and constrained viewpoints. On the modeling side, current approaches predominantly rely on supervised fine-tuning, which yields only marginal improvements in reasoning capabilities. To address these challenges, we propose a novel pipeline that constructs a reasoning-oriented geo-localization dataset, MP16-Reason, using diverse social media images. We introduce GLOBE, Group-relative policy optimization for Localizability assessment and Optimized visual-cue reasoning, yielding Bi-objective geo-Enhancement for the VLM in recognition and reasoning. GLOBE incorporates task-specific rewards that jointly enhance localizability assessment, visual-cue reasoning, and geolocation accuracy. Both qualitative and quantitative results demonstrate that GLOBE outperforms state-of-the-art open-source LVLMs on geo-localization tasks, particularly in diverse visual scenes, while also generating more insightful and interpretable reasoning trajectories. The data and code are available at https://github.com/lingli1996/GLOBE. |
| title | Recognition through Reasoning: Reinforcing Image Geo-localization with Large Vision-Language Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.14674 |