Recently, several methods have been reported for the contact-free measurement of heart rate from human face videos. Typically, these are based on applying Independent Component Analysis (ICA) to sequences of colour values from tracked face patches and selecting the peak value within a fixed range in the frequency spectrum. Those methods performed well enough under controlled conditions, but in realistic situations their performance degraded significantly. In this paper, we have highlighted issues such as light reflection and participant's movement, and proposed solutions to those factors that reduce the performance of existing methods. We experiment with calculating a robust mean of the skin colour pixels values using a mean shift algorithm and estimating an optimal frequency selection function in the Fourier domain. Preliminary results show that these methods can improve the results of existing methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.