The viewer is disabled because this dataset repo requires arbitrary Python code execution. Please consider removing the loading script and relying on automated data support (you can use convert_to_parquet from the datasets library). If this is not possible, please open a discussion for direct help.

image/png

OpenStreetView-5M
The Many Roads to Global Visual Geolocation πŸ“πŸŒ

First authors: Guillaume Astruc, Nicolas Dufour, Ioannis Siglidis
Second authors: Constantin Aronssohn, Nacim Bouia, Stephanie Fu, Romain Loiseau, Van Nguyen Nguyen, Charles Raude, Elliot Vincent, Lintao XU, Hongyu Zhou
Last author: Loic Landrieu
Research Institute: Imagine, LIGM, Ecole des Ponts, Univ Gustave Eiffel, CNRS, Marne-la-VallΓ©e, France

Introduction 🌍

OpenStreetView-5M is the first large-scale open geolocation benchmark of streetview images.
To get a sense of the difficulty of the benchmark, you can play our demo.

Dataset πŸ’Ύ

To download the datataset, run:

# download the full dataset
from huggingface_hub import snapshot_download
snapshot_download(repo_id="osv5m/osv5m", local_dir="datasets/osv5m", repo_type='dataset')

and finally extract:

import os
import zipfile
for root, dirs, files in os.walk("datasets/osv5m"):
    for file in files:
        if file.endswith(".zip"):
            with zipfile.ZipFile(os.path.join(root, file), 'r') as zip_ref:
                zip_ref.extractall(root)
                os.remove(os.path.join(root, file))

You can also directly load the dataset using load_dataset:

from datasets import load_dataset
dataset = load_dataset('osv5m/osv5m', full=False)

where with full you can specify whether you want to load the complete metadata (default: False).

If you only want to download the test set, you can run the script below:

from huggingface_hub import hf_hub_download
for i in range(5):
    hf_hub_download(repo_id="osv5m/osv5m", filename=str(i).zfill(2)+'.zip', subfolder="images/test", repo_type='dataset', local_dir="datasets/OpenWorld")
    hf_hub_download(repo_id="osv5m/osv5m", filename="README.md", repo_type='dataset', local_dir="datasets/OpenWorld")

Citing πŸ’«

@article{osv5m,
    title = {{OpenStreetView-5M}: {T}he Many Roads to Global Visual Geolocation},
    author = {Astruc, Guillaume and Dufour, Nicolas and Siglidis, Ioannis
      and Aronssohn, Constantin and Bouia, Nacim and Fu, Stephanie and Loiseau, Romain
      and Nguyen, Van Nguyen and Raude, Charles and Vincent, Elliot and Xu, Lintao
      and Zhou, Hongyu and Landrieu, Loic},
    journal = {CVPR},
    year = {2024},
  }
Downloads last month
100