DocTrack Dataset

This dataset was created by the Natural Language Understanding and Human-Computer Interaction Laboratory of Shanghai University with the purpose of research on human-like visually-rich document understanding.

Note: The DocTrack dataset should only be used for non-commercial research purposes. For any person/institution/company working on this direction, please contact us for a commercial license.

Description

DocTrack contains 539 images along with their eye-tracking order annotations. The original images are collected from the FUNSD, SEABILL and Inforgraphic VQA datasets. For more details, please refer to our paper accepted by EMNLP2023(findings) DocTrack: A Visually-Rich Document Dataset Really Aligned with Human Eye Movement for Machine Reading.

Citation

@misc{wang2023dc,
    title={DocTrack: A Visually-Rich Document Dataset Really Aligned with Human Eye Movement for Machine Reading},
    author={Hao Wang, Qingxuan Wang, Yue Li, Changqing Wang, Chenhui Chu and Rui Wang},
    year={2023},
    eprint={2310.14802},
    archivePrefix={arXiv},
    primaryClass={cs.HC}
}