LQ-rPPG: A Label-Quantized Coarse-to-Fine Learning Framework for Remote Physiological Measurement

2026-05-21

AI Summary

This work addresses the challenge of noisy and highly variable contact-based photoplethysmography (PPG) labels in remote PPG (rPPG) estimation, which often leads to model overfitting and poor generalization. To mitigate this issue, the authors propose a coarse-to-fine learning framework based on label quantization. The approach introduces, for the first time, a multi-bit pseudo-labeling mechanism that discretizes continuous PPG signals to suppress label noise, coupled with hierarchical supervision to progressively refine rPPG signal estimation. The proposed framework significantly enhances model robustness and cross-domain generalization, achieving state-of-the-art performance across multiple benchmark datasets while substantially reducing computational overhead—parameters and MACs are decreased by 88% and 29%, respectively, and throughput is improved by 191%.

Abstract

Remote photoplethysmography (rPPG) enables non-contact measurement of physiological signals from facial videos, offering strong potential for remote healthcare and daily health monitoring. Driven by this potential, various deep learning-based rPPG methods have been proposed to improve rPPG estimation. However, previous deep learning-based rPPG methods have paid little attention to the quality of training labels and their impact on model learning. Contact-based PPG signals used as training labels often contain noise and variability caused by motion artifacts, inconsistent sensor contact, and morphological distortions. Such label inconsistency can lead models to overfit to the label noise and variability and consequently degrade generalization performance. To address this issue, we propose LQ-rPPG, a label-quantized coarse-to-fine learning framework for robust rPPG estimation. LQ-rPPG consists of a label quantization module and a coarse-to-fine rPPG estimation model. The label quantization module transforms continuous PPG signals into multi-bit quantized pseudo labels with reduced noise and variability. The coarse-to-fine estimation model progressively refines rPPG signals under hierarchical supervision guided by the multi-bit pseudo labels. This design alleviates overfitting to label-specific variations and enables the model to learn structured and consistent representations. As a result, LQ-rPPG achieves robust and generalizable rPPG estimation even under challenging conditions. Experiments on multiple benchmark datasets demonstrate that LQ-rPPG achieves strong performance in both intra- and cross-dataset evaluations, while reducing parameters and multiply-accumulate operations by 88% and 29%, respectively, and increasing throughput by 191%. The code is available at https://github.com/Anonymous-repo-code/LQ-rPPG.

Problem

• remote photoplethysmography
• label noise
• training label quality
• model generalization
• physiological signal estimation

Innovation

• label quantization
• coarse-to-fine learning
• remote photoplethysmography
• pseudo labels
• robust physiological measurement