An Efficient Approach to Estimating Heart Rate from Facial Videos with Accurate Region of Interest

Contactless Heart rate measurement of the cardiac pulses provides an effective way of physical assessment during pandemic situations where physical contact is highly risky. Remote photoplethysmography (RPPG) provides the solution for contactless monitoring of blood volume pulse from facial videos. This article presents an efficient technique to estimating the heart rate from facial regions. The work mainly focuses on extracting the Region of Interest (ROI) from face images which are robust against illumination and motion changes. The work includes the skin standardization mechanism that could be able to locate the appropriate ROI irrespective of the skin tone of the subject, thereby increasing the effectiveness and application of the approach in real-time scenarios. The proposed method compares with traditional methods to assess and evaluate accuracy, consistency, and sensitivity. The proposed approach attains faster processing with a decrease in the number of elements for processing decreased by 5 times.