Set of models for emotion estimation from EEG. Composed by the combination of two deep-learing models learning together (RNN and CNN) with the help of a saliency analysis.
The three proposed models are direcly available here:
- Saliency based combination of the hierarchical RNN and CNN. Sal_Model.py
- Feature fusion of the hierarchical RNN and CNN. Feat_Model.pySal_Model* Ouput fusion of the hierarchical RNN and CNN. Loss_Model.py
Installation with pip: pip install -r req.txt
Import of the environment with conda: conda env create -f env.yml
Due to the EULA for each dataset, some example signals have been proposed to test the models, however, they are not corresponding to signals from one of tested dataset.
If you are interested in our work, don't hesitate to contact us.
Best! 😄
ps: if you use this repo in other research project, please cite the original paper:
@article{delvigne2022emotion,
title={Emotion Estimation from EEG--A Dual Deep Learning Approach Combined with Saliency},
author={Delvigne, Victor and Facchini, Antoine and Wannous, Hazem and Dutoit, Thierry and Ris, Laurence and Vandeborre, Jean-Philippe},
journal={arXiv preprint arXiv:2201.03891},
year={2022}
}