Deep Learning Methods for Non-Contact Heart Rate Monitoring: A Review

Abstract

Deep learning is a branch of machine learning that aims to teach computers to perform tasks that mimic human-like natural abilities. It has emerged as a pivotal technology contributing to advancements in the healthcare industry. This paper gives a thorough summary of the use of deep learning in non-contact heart rate assessment. Heart rate (HR) serves as a crucial vital sign for estimating the physiological condition of the body. Unlike conventional HR monitors, which typically necessitate skin contact, remote photoplethysmography (rPPG) allows for non-contact HR measurement by detecting minute variations in skin illumination through a video camera.