Camera-Based Remote Physiology Sensing for Hundreds of Subjects Across Skin Tones

TL;DR

This paper introduces the VitalVideo dataset, the largest real-world rPPG dataset with diverse skin tones, and demonstrates that datasets with a few hundred subjects are sufficient for effective model training and evaluation.

Contribution

It presents the VitalVideo dataset with 893 subjects across six skin tones and analyzes the impact of dataset size and diversity on rPPG model performance.

Findings

Datasets with 300-700 subjects are sufficient for effective rPPG training.

Diversity in skin tones is crucial for accurate performance evaluation.

Model performance improves with increased dataset size and skin tone representation.

Abstract

Remote photoplethysmography (rPPG) emerges as a promising method for non-invasive, convenient measurement of vital signs, utilizing the widespread presence of cameras. Despite advancements, existing datasets fall short in terms of size and diversity, limiting comprehensive evaluation under diverse conditions. This paper presents an in-depth analysis of the VitalVideo dataset, the largest real-world rPPG dataset to date, encompassing 893 subjects and 6 Fitzpatrick skin tones. Our experimentation with six unsupervised methods and three supervised models demonstrates that datasets comprising a few hundred subjects(i.e., 300 for UBFC-rPPG, 500 for PURE, and 700 for MMPD-Simple) are sufficient for effective rPPG model training. Our findings highlight the importance of diversity and consistency in skin tones for precise performance evaluation across different datasets.