SUMS

Here is SUMS dataset collected by Qinghai University. It is a multi-camera and multi-signal Biosensing dataset collected at high altitudes, which includes 80 synchronized non-contact facial and contact finger videos from 10 subjects during exercise and oxygen recovery scenarios. This dataset captures PPG, respiration rate (RR), and SpO2, and is designed to validate video vitals estimation algorithms and compare facial rPPG with finger cPPG. Our results demonstrate that fusing videos from different positions (face and finger) reduces the mean absolute error (MAE) of SpO2 predictions by 7.6% and 10.6% compared to using only face or only finger data. Additionally, training on multiple indicators such as PPG and blood oxygen simultaneously reduces SpO2 estimation MAE by 17.8%. We recruited ten participants living on the Qinghai Plateau to collect hypoxia data in a real high-altitude environment. Data collection utilized two Logitech C922 cameras to capture videos of participants’ faces and fingers, along with a CMS50E+ oximeter for measuring PPG signals and blood oxygen levels. Respiratory patterns were monitored using an HKH-11C respiratory sensor. Video recordings were captured at a resolution of 1280x720 pixels, running at 60 frames per second. The PPG signals were recorded at a frequency of 20 Hz, while respiratory waves were captured at 50 Hz. The experiment setup is shown as follows.