Proposed Algorithm Overview

Abstract

Inter-beat interval (IBI) and heart rate variability (HRV) are important cardiac parameters that provide physiological and emotional states of a person. In this paper, we present a framework for accurate IBI and HRV estimation from a facial video based on the reliability of extracted blood volume pulse (BVP) signals. Our framework first extracts candidate BVP signals from randomly sampled multiple face patches. The BVP signals are then assessed based on a reliability metric to select the most reliable BVP signal, from which IBI and HRV are calculated. In experiments, we evaluate three reliability metrics and demonstrate that our framework can estimate IBI and HRV more accurately than a conventional single face region-based framework.

Publications

• Inter-Beat Interval Estimation from Facial Video Based on Reliability of BVP Signals [PDF] [Slide]
  Yuichiro Maki, Yusuke Monno, Kazunori Yoshizaki, Masayuki Tanaka, Masatoshi Okutomi
  International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC2019), pp.6525-6528, July 2019.

Dataset and Code

• Dataset [TokyoTech Remote PPG Dataset]
  This dataset contains facial videos of 9 subjects and reference contact PPG sensor data, which were used for the evaluation of IBI estimation in our EMBC paper.
• Code [Github]
  This code provides MATLAB implementation of our proposed framework, which estimates IBI (or mean heart rate) from a facial video based on the reliability of BVP signals.