Pulse rate estimation using imaging photoplethysmography: generic framework and comparison of methods on a publicly available dataset

TL;DR

This paper establishes a comprehensive framework and benchmark dataset for comparing pulse rate estimation methods using imaging photoplethysmography (iPPG) from facial videos, demonstrating the effectiveness of the POS method combined with continuous wavelet transform.

Contribution

It introduces a generic framework for comparing iPPG pulse rate estimation methods and provides benchmark results on a publicly available dataset.

Findings

Best accuracy achieved with POS method and continuous wavelet transform

Overall mean absolute error below 2 beats per minute

Benchmark dataset enables standardized comparison of methods

Abstract

Objective: to establish an algorithmic framework and a benchmark dataset for comparing methods of pulse rate estimation using imaging photoplethysmography (iPPG). Approach: first we reveal essential steps of pulse rate estimation from facial video and review methods applied at each of the steps. Then we investigate performance of these methods for DEAP dataset www.eecs.qmul.ac.uk/mmv/datasets/deap/ containing facial videos and reference contact photoplethysmograms. Main results: best assessment precision is achieved when pulse rate is estimated using continuous wavelet transform from iPPG extracted by the POS method (overall mean absolute error below 2 heart beats per minute). Significance: we provide a generic framework for theoretical comparison of methods for pulse rate estimation from iPPG and report results for the most popular methods on a publicly available dataset that can be used as a benchmark.