README.md

metadata

configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
task_categories:
  - visual-question-answering
language:
  - fra
license: cc-by-4.0

Description

This dataset is a processed version of the Localized Narratives dataset by Pont-Tuset et al., particularly for a visual question answering task where answer is a caption.
We converted the images to PIL format (image column of the dataset); created 64 different questions in French (question column); and finally translated the original captions from English to French and used them as answers (answer column).

Citation

Localized Narratives

@misc{ponttuset2020connectingvisionlanguagelocalized,
      title={Connecting Vision and Language with Localized Narratives}, 
      author={Jordi Pont-Tuset and Jasper Uijlings and Soravit Changpinyo and Radu Soricut and Vittorio Ferrari},
      year={2020},
      eprint={1912.03098},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/1912.03098}, 
}