randall-lab/INRIA-holidays

Dataset Card for INRIA Holidays

Citation

Per INRIA, if you use this dataset, please cite the following paper:

Herve Jegou, Matthijs Douze, and Cordelia Schmid
"Hamming Embedding and Weak Geometry Consistency for Large Scale Image Search"
Proceedings of the 10th European Conference on Computer Vision, October 2008.

Dataset Details

Description

The INRIA Holidays dataset is a benchmark dataset for large-scale image retrieval tasks. It contains a collection of personal holiday photos grouped by scene, designed to evaluate the performance of image retrieval algorithms under realistic conditions.

The INRIA Holidays dataset contains a total of 1,491 images grouped into 500 scenes, where each scene includes one query image and multiple similar images.

• Data components:
  • scene_id — Identifier string for the scene group of images. This is a number.
  • query — A query image from the scene.
  • similar — A list of images similar to the query image within the same scene.

The dataset supports retrieval tasks where given a query image, the goal is to find all visually similar images in the dataset.

Dataset Source

• Homepage: INRIA Holidays Dataset
• Paper:
  Herve Jegou, Matthijs Douze, Cordelia Schmid,
  Hamming Embedding and Weak Geometry Consistency for Large Scale Image Search,
  European Conference on Computer Vision (ECCV), 2008.

This uses a Kaggle Version to pull the data.

Example Usage

You need to have kagglehub to run this download:

pip install kagglehub

from datasets import load_dataset

# Load the INRIA Holidays dataset
ds = load_dataset("randall-lab/INRIA-holidays", split="train", trust_remote_code=True)

# Access a sample
example = ds[0]
print("Scene ID:", example["scene_id"])

# Show query image
example["query"].show()

# Show first similar image
example["similar"][0].show()

License and Disclaimer

This dataset is provided "as is" by INRIA without any warranties or guarantees. Users assume all responsibility for how they use the dataset.

Contact and Support

For questions or issues regarding the dataset, please refer to the original INRIA Holidays dataset webpage or contact INRIA.

Citation in BibTeX format

@inproceedings{jegou2008hamming,
  title={Hamming embedding and weak geometry consistency for large scale image search},
  author={Jegou, Herve and Douze, Matthijs and Schmid, Cordelia},
  booktitle={European conference on computer vision},
  pages={304--317},
  year={2008},
  organization={Springer}
}