Remote Photoplethysmography

Remote photoplethysmography (rPPG) is a non-contact technique for measuring physiological signals, primarily heart rate, from facial videos using a camera. Current research focuses on improving the robustness and generalizability of rPPG models across diverse conditions (lighting, skin tones, motion) and datasets, employing various deep learning architectures including convolutional neural networks (CNNs), transformers, and state-space models like Mamba, often incorporating techniques like test-time adaptation and domain adaptation. These advancements are significant for expanding the applications of rPPG in healthcare monitoring, affective computing, and biometric authentication, offering a convenient and privacy-sensitive alternative to contact-based methods.