metadata
dataset_info:
features:
- name: obs_uid
dtype: string
- name: usr_uid
dtype: string
- name: caption
dtype: string
- name: image
dtype: image
- name: clicks_path
sequence:
sequence: int32
length: 2
- name: clicks_time
sequence: timestamp[s]
splits:
- name: train
num_bytes: 1611467
num_examples: 3848
download_size: 241443505
dataset_size: 1611467
Dataset Description
CapMIT1003 is a dataset of captions and click-contingent image explorations collected during captioning tasks. CapMIT1003 is based on the same stimuli from the well-known MIT1003 benchmark, for which eye-tracking data under free-viewing conditions is available, which offers a promising opportunity to concurrently study human attention under both tasks.
Usage
You can load CapMIT1003 as follows:
from datasets import load_dataset
capmit1003_dataset = load_dataset("azugarini/CapMIT1003", trust_remote_code=True)
print(capmit1003_dataset["train"][0]) #print first example
Citation Information
If you use this dataset in your research or work, please cite the following paper:
@article{zanca2023contrastive,
title={Contrastive Language-Image Pretrained Models are Zero-Shot Human Scanpath Predictors},
author={Zanca, Dario and Zugarini, Andrea and Dietz, Simon and Altstidl, Thomas R and Ndjeuha, Mark A Turban and Schwinn, Leo and Eskofier, Bjoern},
journal={arXiv preprint arXiv:2305.12380},
year={2023}