LLMGeo: Benchmarking Large Language Models on Image Geolocation In-the-wild
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911895277010944 |
|---|---|
| author | Wang, Zhiqiang Xu, Dejia Khan, Rana Muhammad Shahroz Lin, Yanbin Fan, Zhiwen Zhu, Xingquan |
| author_facet | Wang, Zhiqiang Xu, Dejia Khan, Rana Muhammad Shahroz Lin, Yanbin Fan, Zhiwen Zhu, Xingquan |
| contents | Image geolocation is a critical task in various image-understanding applications. However, existing methods often fail when analyzing challenging, in-the-wild images. Inspired by the exceptional background knowledge of multimodal language models, we systematically evaluate their geolocation capabilities using a novel image dataset and a comprehensive evaluation framework. We first collect images from various countries via Google Street View. Then, we conduct training-free and training-based evaluations on closed-source and open-source multi-modal language models. we conduct both training-free and training-based evaluations on closed-source and open-source multimodal language models. Our findings indicate that closed-source models demonstrate superior geolocation abilities, while open-source models can achieve comparable performance through fine-tuning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_20363 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | LLMGeo: Benchmarking Large Language Models on Image Geolocation In-the-wild Wang, Zhiqiang Xu, Dejia Khan, Rana Muhammad Shahroz Lin, Yanbin Fan, Zhiwen Zhu, Xingquan Computer Vision and Pattern Recognition Image geolocation is a critical task in various image-understanding applications. However, existing methods often fail when analyzing challenging, in-the-wild images. Inspired by the exceptional background knowledge of multimodal language models, we systematically evaluate their geolocation capabilities using a novel image dataset and a comprehensive evaluation framework. We first collect images from various countries via Google Street View. Then, we conduct training-free and training-based evaluations on closed-source and open-source multi-modal language models. we conduct both training-free and training-based evaluations on closed-source and open-source multimodal language models. Our findings indicate that closed-source models demonstrate superior geolocation abilities, while open-source models can achieve comparable performance through fine-tuning. |
| title | LLMGeo: Benchmarking Large Language Models on Image Geolocation In-the-wild |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2405.20363 |