Comparative Analysis of rPPG Algorithms Across Diverse Countries of Origin

Heart rate measurement is crucial for assessing physical health. Traditionally, contact-based method such as PPG requiring skin attachment have been predominant. However, these methods have limitations due to their invasive nature. Remote photoplethysmography (rPPG) offers a non-contact alternative that broadens its application possibilities. Yet, current rPPG methods have not been extensively tested across diverse populations, where varying skin pigmentation can affect measurement accuracy. This study aims to determine the most effective rPPG method for extracting heart rate from individuals of diverse origins. Evaluating Chrominance-based (CHROM), Plane Orthogonal to Skin (POS), and GREEN methods using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), we found that the POS method demonstrated superior predictive accuracy with an RMSE of 8.45, while GREEN showed comparable performance with an RMSE of 8.89 across different country groups.