GOO, short for Gaze-on-Objects, is a gaze object prediction dataset for retail environments. The goal of this dataset is to predict the bounding box of an object being gazed at. GOO consists of a large set of synthetic images (GOO Synth) and a smaller subset of real images (GOO-Real), the latter of which are images of people looking at objects in retail environments.

Citation

@inproceedings{tomas2021goo,
title={GOO: A Dataset for Gaze Object Prediction in Retail Environments},
author={Tomas, Henri and Reyes, Marcus and Dionido, Raimarc and Ty, Mark and Casimiro, Joel and Atienza, Rowel and Guinto, Richard},
booktitle = {CVPR Workshops (CVPRW)},
year={2021},
pubstate={published},
tppubtype={inproceedings}
}