Gaze into the Heart: A Multi-View Video Dataset for rPPG and Health Biomarkers Estimation
Konstantin Egorov, Stepan Botman, Pavel Blinov, Galina Zubkova, Anton Ivaschenko, Alexander Kolsanov, Andrey Savchenko
2025-08-28
Summary
This paper introduces a new, large collection of video and physiological data designed to help researchers improve remote health monitoring using just a camera.
What's the problem?
Currently, it's hard to develop and test reliable remote health monitoring systems because the existing datasets are too small, raise privacy issues with facial videos, and don't include enough variety in the conditions people are in when the data is collected. This limits progress in creating accurate AI tools for health.
What's the solution?
The researchers created a dataset with videos of 600 people recorded with multiple cameras from different angles, both while resting and after exercise. Importantly, they paired these videos with direct measurements of vital signs like heart rate, blood pressure, oxygen levels, and even stress levels. They also used this data to build and test a new rPPG model, comparing its performance to existing methods.
Why it matters?
By making this dataset and their model publicly available, the researchers hope to accelerate the development of AI-powered medical assistants that can monitor your health remotely using just a camera, potentially leading to earlier detection of health problems and more convenient healthcare.
Abstract
Progress in remote PhotoPlethysmoGraphy (rPPG) is limited by the critical issues of existing publicly available datasets: small size, privacy concerns with facial videos, and lack of diversity in conditions. The paper introduces a novel comprehensive large-scale multi-view video dataset for rPPG and health biomarkers estimation. Our dataset comprises 3600 synchronized video recordings from 600 subjects, captured under varied conditions (resting and post-exercise) using multiple consumer-grade cameras at different angles. To enable multimodal analysis of physiological states, each recording is paired with a 100 Hz PPG signal and extended health metrics, such as electrocardiogram, arterial blood pressure, biomarkers, temperature, oxygen saturation, respiratory rate, and stress level. Using this data, we train an efficient rPPG model and compare its quality with existing approaches in cross-dataset scenarios. The public release of our dataset and model should significantly speed up the progress in the development of AI medical assistants.