osv5m/osv5m
收藏资源简介:
OpenStreetView-5M是第一个大规模开放的街景图像地理定位基准数据集。该数据集旨在提供一个全球范围内的视觉地理定位挑战,用户可以通过提供的演示来感受基准的难度。数据集包含训练集和测试集,用户可以通过Hugging Face Hub下载完整数据集或仅下载测试集,并提供了加载数据集的代码示例。
OpenStreetView-5M is the first large-scale open street view image geolocation benchmark dataset. This dataset aims to provide a global visual geolocation challenge, where users can experience the difficulty of the benchmark through the provided demonstrations. The dataset includes a training set and a test set. Users can download either the full dataset or only the test set via the Hugging Face Hub, and code examples for loading the dataset are also provided.
OpenStreetView-5M
数据集概述
OpenStreetView-5M 是一个大规模的开放式街景图像地理定位基准数据集。
数据集结构
- 配置名称: default
- 数据文件:
- 训练集:
- 文件路径: "train.csv"
- 图像路径: "images/train"
- 测试集:
- 文件路径: "test.csv"
- 图像路径: "images/test"
- 训练集:
数据集下载
下载完整数据集
python from huggingface_hub import snapshot_download snapshot_download(repo_id="osv5m/osv5m", local_dir="datasets/osv5m", repo_type=dataset)
解压数据集
python 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))
直接加载数据集
python from datasets import load_dataset dataset = load_dataset(osv5m/osv5m, full=False)
其中 full 参数用于指定是否加载完整元数据(默认值为 False)。
仅下载测试集
python 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")
引用
bibtex @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}, }




