Konstantin Egorov Stepan Botman Pavel Blinov Galina Zubkova Anton Ivaschenko Alexander Kolsanov Andrey Savchenko

Abstract

Progress in remote PhotoPlethysmoGraphy (rPPG) is limited by the criticalissues of existing publicly available datasets: small size, privacy concernswith facial videos, and lack of diversity in conditions. The paper introduces anovel comprehensive large-scale multi-view video dataset for rPPG and healthbiomarkers estimation. Our dataset comprises 3600 synchronized video recordingsfrom 600 subjects, captured under varied conditions (resting and post-exercise)using multiple consumer-grade cameras at different angles. To enable multimodalanalysis of physiological states, each recording is paired with a 100 Hz PPGsignal and extended health metrics, such as electrocardiogram, arterial bloodpressure, biomarkers, temperature, oxygen saturation, respiratory rate, andstress level. Using this data, we train an efficient rPPG model and compare itsquality with existing approaches in cross-dataset scenarios. The public releaseof our dataset and model should significantly speed up the progress in thedevelopment of AI medical assistants.