MVPD: A Multimodal Video Physiology Database for rPPG

During the era of rapid development in the field of computer vision, camera sensors have been applied to a variety of scenarios, able to capture states and movement changes of objects, and even observe nuances in the real world that are invisible to human eyes. Remote photoplethysmography (rPPG) is a camera-based non-contact sensor method for monitoring physiological signs. In recent years, there are few datasets for rPPG and most of them are only for a few physiological parameters, such as photoplethysmography (PPG), heart rate (HR) and peripheral oxygen saturation (SpO2). For studying more vital signs and learnable parameters, we propose the Multimodal Video Physiology Database (MVPD) containing videos and physiological information from 13 volunteers. We use cameras to collect videos of face, palm, wrist and fingertip, and physiological sign parameters， such as electrocardiogram (ECG), respiratory rate (RR), skin temperature (ST), PPG, pulse rate (PR), blood pressure (BP), and SpO2. This database is mainly used for studying rPPG.