hint-lab/doctrack

Dataset for EMNLP'23 Paper "DocTrack: A Visually-Rich Document Dataset Really Aligned with Human Eye Movement for Machine Reading"

28
/ 100
Experimental

This dataset offers 539 document images paired with human eye-tracking data, showing the order in which a person visually processes information on a page. It provides researchers with real human eye movement patterns on documents, helping to develop systems that understand documents more like humans do. Scientists studying how people read and process visual information in documents would use this.

No commits in the last 6 months.

Use this if you are conducting academic research on how humans read visually complex documents and want to train AI models to mimic that understanding.

Not ideal if you require a dataset for commercial applications, as this is restricted to non-commercial research use only.

human-computer-interaction document-understanding-research eye-tracking-studies reading-comprehension-AI cognitive-science
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

11

Forks

1

Language

License

Apache-2.0

Last pushed

Oct 25, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/hint-lab/doctrack"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.