Global visual salience of competing stimuli

By Alex Hernandez-Garcia, Ricardo Ramos Gameiro, Alessandro Grillini and Peter König

This repository contains the code used to compute and analyse the global visual salience of natural images from eye-tracking data, as described in the paper Global visual salience of competing stimuli (Journal of Vision, 2020).

Data

The complete data sets of this project are available on the Supplementary material OSF page of the preprint. The following files are available:

• data_raw.mat: The basic data set containing the eye-tracking data from the experimental sessions.
• data_all.csv: The most complete version of the data set, containing the original data from data_raw.mat, as well as other useful information.
• data_firstfixation.csv: A derived data set containing data relative to the first fixations at each trial. Also available in this repository (./data/data_firstfixation.csv)

Usage

1. Clone the repository:

git clone https://github.com/alexhernandezgarcia/global-salience.git

2. Install the package

pip

Run the following command from inside the directory:

pip install -e .

3.1 Try some of the examples

python ./examples/eval.py --input ./data/data_firstfixation.csv --target first --test_pct 0.2 --test_folds 25

3.2 Analyze the data as in the Jupyter notebooks

Citation

If you use this code for scientific purposes, please cite:

Alex Hernandez-Garcia, Ricardo Ramos Gameiro, Alessandro Grillini, Peter König, 2019. Global visual salience of competing stimuli. PsyArXiv:z7qp5

@article{hergar2019globalsalience,
	author = {Hernandez-Garcia, Alex and Ramos Gameiro, Ricardo and Grillini, Alessandro and K{\"o}nig, Peter},
	title = {Global visual salience of competing stimuli},
	journal = {Journal of Vision},
	year = {2020},
	volume = {20},
	number = {7},
	pages = {27-27},
	month = {07},
	doi = {10.1167/jov.20.7.27},
}

License

Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License