A Skin-Tone-Aware Dual-Representation Remote Photoplethysmography Framework for Contactless Respiratory Rate Estimation

2026-06-19

AI Summary

This study addresses the limitations of existing non-contact respiratory rate (RR) estimation methods, which often repurpose heart rate algorithms and struggle to model respiratory dynamics while being susceptible to skin tone variations and non-respiratory motion artifacts. To overcome these challenges, this work proposes a dual-representation framework that fuses a skin-tone-aware Eulerian representation with a denoised Lagrangian representation, enabling robust RR estimation through dynamic RGB signal projection and motion artifact suppression. Key innovations include a Lagrangian denoising network tailored for respiratory signals, a phase-invariant contrastive learning loss, and a dynamic skin-tone-adaptive projection mechanism. The authors also introduce RR-rPPG, the first dataset incorporating Indian subjects for remote photoplethysmography-based RR estimation. Experiments demonstrate significant performance gains over state-of-the-art methods on both RR-rPPG and COHFACE, achieving up to a 42.1% reduction in mean absolute error. The code and dataset are publicly released.

Abstract

Respiratory rate is a vital indicator of pulmonary and cardiovascular health, yet conventional methods for estimating respiratory rate are often intrusive due to their contact-based nature. Remote photoplethysmography offers a promising non-contact alternative and has been widely used for heart rate estimation; however, its potential for respiratory rate estimation remains underexplored. Existing methods typically adapt green and chrominance-based projections originally designed for heart rate estimation, which only partially capture respiratory dynamics. Most prior work focuses on the Eulerian representation with fixed or empirically selected RGB projections. To address these gaps, we propose a skin-tone-aware dynamic RGB signal projection that captures respiratory information. To mitigate the sensitivity of the Lagrangian representation to non-respiratory motion, we introduce a denoising network for motion-based remote photoplethysmography signals. We further design a phase-independent contrastive loss that enables Eulerian and Lagrangian representations to collaboratively learn respiratory rate information. We also introduce RR-rPPG, a respiratory-rate facial video dataset with Indian demographic representation. We evaluate the method on RR-rPPG and the publicly available COHFACE dataset, where it consistently outperforms comparison methods and achieves up to a 42.1% reduction in mean absolute error across the evaluated settings. The proposed framework demonstrates the effectiveness of jointly leveraging skin-tone-aware Eulerian and denoised Lagrangian representations for contactless respiratory rate estimation from facial videos. In addition, RR-rPPG contributes a diverse benchmark resource for future research in remote respiratory monitoring. The code and dataset will be made publicly available upon paper acceptance.

Problem

• remote photoplethysmography
• respiratory rate estimation
• non-contact monitoring
• skin tone bias
• motion artifact

Innovation

• remote photoplethysmography
• respiratory rate estimation
• skin-tone-aware projection
• dual-representation learning
• motion denoising