A Plug-and-Play Temporal Normalization Module for Robust Remote Photoplethysmography

TL;DR

This paper presents a novel Temporal Normalization module that enhances remote photoplethysmography accuracy by effectively mitigating motion and lighting artifacts, significantly improving performance across multiple models and datasets.

Contribution

Introduces a flexible plug-and-play Temporal Normalization module that captures long-term features, boosting rPPG accuracy and robustness against artifacts in various architectures.

Findings

Performance improvements of 34.3% to 94.2% in heart rate measurement.

Greater gains observed in smaller models.

Effective mitigation of motion and lighting artifacts.

Abstract

Remote photoplethysmography (rPPG) extracts PPG signals from subtle color changes in facial videos, showing strong potential for health applications. However, most rPPG methods rely on intensity differences between consecutive frames, missing long-term signal variations affected by motion or lighting artifacts, which reduces accuracy. This paper introduces Temporal Normalization (TN), a flexible plug-and-play module compatible with any end-to-end rPPG network architecture. By capturing long-term temporally normalized features following detrending, TN effectively mitigates motion and lighting artifacts, significantly boosting the rPPG prediction performance. When integrated into four state-of-the-art rPPG methods, TN delivered performance improvements ranging from 34.3% to 94.2% in heart rate measurement tasks across four widely-used datasets. Notably, TN showed even greater performance gains in smaller models. We further discuss and provide insights into the mechanisms behind TN's effectiveness.