StressID is a Dataset specifically designed for stress identification based on multimodal data collection.

StressID contains video, audio and physiological signals.

It is contains RGB facial video, audio and physiological signals (ECG, EDA, Respiration). Different stress-inducing stimuli are used: emotional video-clips, cognitive tasks and public speaking. The total dataset is consists of recordings from 65 participants that performed 11 tasks. Each task is labeled by the subjects in terms of stress, relaxation, arousal, and valence. The experimental set-up ensures synchronised, high-quality, and low noise data. Code for unimodal and multimodal baseline models is available at https://github.com/robustml-eurecom/stressID.

StressID data and set-up.

CREDITS

We thank all the 65 participant. Without them this dataset and the related research would have never been possible.

We thanks our institutions, INRIA and EURECOM, and the French government that supported us through the following programs: the 3IA Côte d’Azur Investments in the Future project managed by the National Research Agency (ANR) (ANR-19-P3IA-0002), the ANR RESPECT Project (ANR-18-CE92-0024), the UCAJEDI Investments in the Future project managed by Ville de Nice and ANR (ANR-15-IDEX-01) and by PNRR- Investment 1.5 Ecosystems of Innovation, Project Tuscany Health Ecosystem (THE), Spoke
3 “Advanced technologies, methods, materials and heath analytics” CUP: I53C22000780001.

CITATION

@inproceedings{
chaptoukaev2023stressid,
title={Stress{ID}: a Multimodal Dataset for Stress Identification},
author={Hava Chaptoukaev and Valeriya Strizhkova and Michele Panariello and Bianca Dalpaos and Aglind Reka and Valeria Manera and Susanne Thummler and Esma ISMAILOVA and Nicholas W. and Francois Bremond and Massimiliano Todisco and Maria A Zuluaga and Laura M. Ferrari},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track},
year={2023},
url={https://openreview.net/forum?id=qWsQi9DGJb}
}

The poster and the full paper is available at https://nips.cc/virtual/2023/poster/73454

More technical information can be found here

HOW TO DOWNLOAD

Please download, read, fill up the form below, and send it to stressid.dataset@inria.fr.

Please note that only a person with a permanent position in academics can sign the license.

Please use your professional email address in the license and double-check it. Please attach to your request a link to a webpage that shows the affiliation and the professional email address.

You are asked to insert your information at PAG.7, 8 and 10.

The Dataset is licensed for non-commercial scientific research purposes.

Any action that violates the license agreement is prosecutable by law.

If you do not receive a feedback in 3 days, please contact laura.ferrari@inria.fr or maria.zuluaga@eurecom.fr

Once validated you will receive a mail with a secured link to download the dataset.

LICENSE.pdf

Comments are closed.