ReViSe: Remote Vital Signs Measurement Using Smartphone Camera

Remote Photoplethysmography (rPPG) is a fast, effective, inexpensive and convenient method for collecting biometric data as it enables vital signs estimation using face videos. contactless medical service provisioning has proven to be dire necessity during the COVID-19 pandemic. We propose an end-to-end framework measure people’s including Heart Rate (HR), Variability (HRV), Oxygen Saturation (SpO2) Blood Pressure (BP) based on rPPG methodology from video of user’s captured with smartphone camera. extract landmarks deep learning-based neural network model in real-time. Multiple patches also called Regions-of-Interest (RoIs) are extracted by predicted landmarks. Several filters applied reduce noise RoIs cardiac signals Volume Pulse (BVP) signal. The measurements HR, HRV SpO2 validated two public datasets namely TokyoTech Detection (PURE) datasets, which our models achieved following Mean Absolute Errors (MAE): a) 1.73Beats-Per-Minute (bpm) 3.95bpm respectively; b) HRV, 18.55ms 25.03ms respectively, c) SpO2, MAE 1.64% PURE dataset. framework, ReViSe, daily living environment, thereby created Video-HR Our HR 2.49bpm this Since no publicly available existed BP measurement videos, we used dataset fingertip sensor train own dataset, Video-BP. On Video-BP 6.7mmHg Systolic (SBP), 9.6mmHg Diastolic (DBP). ReViSe been videos recorded environment opposed less noisy laboratory reported most state-of-the-art techniques.