Remote Heart Rate

Remote heart rate (RHR) monitoring, using video analysis of facial blood flow (remote photoplethysmography or rPPG), aims to provide a non-contact method for continuous heart rate assessment. Current research focuses on improving the robustness of rPPG algorithms against noise and artifacts from varying lighting, motion, skin tones, and video compression, often employing deep learning models like convolutional neural networks (CNNs) and transformers, sometimes incorporating techniques like normalizing flows for data augmentation or time-frequency analysis for signal processing. These advancements hold significant promise for applications in remote healthcare monitoring, improving the accessibility and convenience of vital sign tracking.