Published December 18, 2025 | Version v1
Dataset Restricted

UL-DD: A multimodal drowsiness dataset using video, biometric, and behavioral data

Authors/Creators

  • 1. EDMO icon University of Louisiana at Lafayette
  • 2. ROR icon Tampere University
  • 3. ROR icon Tampere University of Technology

Description

We present a comprehensive public dataset for driver drowsiness detection, integrating multimodal signals of facial, behavioral, and biometric indicators. Our dataset includes 3D facial video, infrared footage, posture videos, and biometric signals like heart rate, electrodermal activity, blood oxygen saturation, skin temperature, and accelerometer data. This data set provides grip sensor data and telemetry data to provide more information about drivers’ behavior while they are alert and drowsy. Drowsiness levels were self-reported every four minutes using the Karolinska Sleepiness Scale (KSS). Data were collected from 19 subjects in two conditions: when they were fully alert and when they exhibited signs of sleepiness. Unlike other datasets, our multimodal dataset has a continuous duration of 40 minutes for each data collection session per subject, contributing to a total length of 1,400 minutes. We recorded gradual changes in the driver state rather than discrete alert/drowsy labels. This study aims to create a publicly available multimodal dataset of driver drowsiness that captures a wider range of physiological, behavioral, and driving-related signals.

Files

Restricted

The record is publicly accessible, but files are restricted to users with access.

Request access

If you would like to request access to these files, please fill out the form below.

You need to satisfy these conditions in order for this request to be accepted:

ACCESS REQUEST INSTRUCTIONS

The UL-DD dataset is available under restricted access to ensure data privacy and proper academic use. To request access, your application must meet the following criteria. Requests failing to meet these requirements will be automatically declined:

  1. Registered Account: You must request access using a registered Zenodo account. Guest requests are not accepted.

  2. Institutional Email: Your request must be associated with a valid institutional, university, or corporate R&D email address. Requests from personal email providers will not be approved.

  3. Statement of Purpose: In your request message, please provide a brief description of your intended research and your institutional affiliation.

  4. Explicit Consent: You must include the following exact statement in your request message: "I have read and agree to abide by the Data Usage Agreement for the UL-DD Dataset."

----------------------------------------
DATA USAGE AGREEMENT
Research Use License for UL-DD Dataset

1. Purpose

This agreement sets forth the terms and conditions for accessing and using the UL-DD dataset, which is provided exclusively for academic and research purposes.

2. Permitted Use

The dataset may be used for any bona fide research purpose, including academic, clinical, or commercial R & D, so long as users remain within the bounds of this agreement and applicable law.

3. Prohibited Actions

Users of the dataset agree to:

  • Not to republish, distribute, or share the dataset in whole or in part outside the intended research scope.
  • Not to use the dataset for advertising purposes.

It is hereby agreed between the data requester, hereinafter referred to as the "LICENSEE", and the University of Louisiana at Lafayette:

  • The LICENSEE will not attempt to identify any individual referenced in the restricted data.
  • The LICENSEE will exercise all reasonable and prudent care to avoid disclosure of the identity of any individual referenced in the restricted data in any publication or other communication
  • The LICENSEE will not share access to the restricted data with anyone else.
  • The LICENSEE will exercise all reasonable and prudent care to maintain the physical and electronic security of the restricted data.
  • If the LICENSEE finds information within the restricted data that they believe might permit identification of any individual, the LICENSEE will report the location of this information promptly by email to CAAI@louisiana.edu, citing the location of the specific information in question.
  • The LICENSEE will use the data for the sole purpose of lawful use in scientific research and no other.
  • The LICENSEE agrees to contribute code associated with publications arising from this data to a repository that is open to the research community.
  • This agreement may be terminated by either party at any time, but the LICENSEE's obligations with respect to the data shall continue after termination.

4. Citation Requirement

 Any publication or research output using the dataset must properly cite the original source, as indicated by the dataset provider.

5. Data Security & Confidentiality

Users must take reasonable precautions to safeguard the dataset from unauthorized access or distribution.

6. Violation Consequences

Failure to comply with these terms may result in revocation of access, legal action, or other corrective measures deemed necessary by the dataset provider.

7. Agreement Acceptance

By downloading or accessing the dataset, users acknowledge and agree to abide by the terms outlined in this agreement.

You are currently not logged in. Do you have an account? Log in here

Additional details

Funding

U.S. National Science Foundation
Improving Drowsiness and Fatigue Prediction of Drivers with Multi-modal Representation Learning and Information Fusion: Application to Traffic Safety 10a.005.UL_TAU