Dataset comprising extracted R, G, and B components for assessment of remote photopletismography

Abstract

This repository contains raw Red (R), Green (G), and Blue (B) components for remote assessment of Pulse Rate (PR). The results of analysis of video signals for noncontact PR assessment are presented in a paper titled "Contactless Pulse Rate Assessment: Results and Insights for Application in Driving Simulators" and published in Applied Sciences journal in 2025. File "RGB dataset.7z" contains 79 compressed .csv (comma-separated values) files for different video files obtained. Each .csv file contains the following columns: Timestamp_HR - timestamp in the UNIX format of the Heart Rate (HR) measurement from Empatica E4 device HR - HR measured at the time indicated by Timestamp_HR by Empatica E4 device in beats per minute (bpm) units Timestamp_BVP - timestamp in the UNIX format of the Blood Volume Pulse (BVP) signal from Empatica E4 device BVP - instantaneous value of the BVP signal, centered around zero and measured using Empatica E4 device (there is no unit for BVP, it is gien in relative signal amplitude) Timestamp_RGB - timestamp of the RGB values assessed from the facial videos in seconds R - average intensity value of the R component from the facial region of interest at the Timestamp_RGB (there is no unit for R component) G - G component at the Timestamp_RGB B - B component at the Timestamp_RGB Acknowledgments: We express our deep gratitude to Jelena Medarević, from the Faculty of Electrical Engineering, University of Ljubljana for her valuable ideas regarding the analysis of signals from the Empatica E4 sensor and for providing us with additional data for assessing Empatica E4 bias (we cite her paper). During the preparation of this study, the first author utilized GPT3.5 (ChatGPT) to enhance clarity and language and the first author is fully responsible for the publication content after using this tool. Funding: Nadica Miljković was financially supported by the Ministry of Science, Technological Development and Innovation of the Republic of Serbia under contract No. 451-03-137/2025-03/200103. The work presented in this paper was financially supported by the Slovenian Research Agency within program ICT4QL, grant no. P2-0246 for Kristina Stojmenova Pečečnik and Jaka Sodnik. Citing Instructions: If you find these data useful for your lecture(s) and/or research, please cite both journal paper and Zenodo repository as: Nešković ĐD, Stojmenova Pečečnik K, Sodnik J, Miljković N. Dataset comprising extracted R, G, and B components for assessment of remote photopletismography (Version 1) [Data set]. Zenodo. 2025. https://doi.org/10.5281/zenodo.16414188 Nešković ĐD, Stojmenova Pečečnik K, Sodnik J, Miljković N. Contactless Pulse Rate Assessment: Results and Insights for Application in Driving Simulators. Applied Sciences. 2025; 15(17):9512. https://doi.org/10.3390/app15179512