Video-based Heart Rate Estimation with Angle-guided ROI Optimization and Graph Signal Denoising

2026-04-13

AI Summary

This study addresses the significant performance degradation of non-contact video-based heart rate estimation under facial motion, such as speaking or head movements. To mitigate motion-induced artifacts, the authors propose two plug-and-play modules: an angle-guided adaptive region-of-interest (ROI) optimization module that corrects motion effects by quantifying the angle between the ROI and the camera, and a multi-region joint graph signal denoising module that leverages graph signal processing to jointly model intra- and inter-regional signals for enhanced motion artifact suppression. This work is the first to integrate angle-guided ROI adaptation with graph-based multi-region denoising. Evaluated on three public datasets, the method reduces the mean absolute error (MAE) by 20.38% compared to baseline approaches, and ablation studies confirm the effectiveness of each component, substantially improving the robustness of remote photoplethysmography (rPPG) in dynamic scenarios.

Technology Category

• Computer Vision: Motion & Tracking
• Search and Optimization: Non-convex Optimization
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Remote photoplethysmography (rPPG) enables non-contact heart rate measurement from facial videos, but its performance is significantly degraded by facial motions such as speaking and head shaking. To address this issue, we propose two plug-and-play modules. The Angle-guided ROI Adaptive Optimization module quantifies ROI-Camera angles to refine motion-affected signals and capture global motion, while the Multi-region Joint Graph Signal Denoising module jointly models intra- and inter-regional ROI signals using graph signal processing to suppress motion artifacts. The modules are compatible with reflection model-based rPPG methods and validated on three public datasets. Results show that jointly use markedly reduces MAE, with an average decrease of 20.38\% over the baseline, while ablation studies confirm the effectiveness of each module. The work demonstrates the potential of angle-guided optimization and graph-based denoising to enhance rPPG performance in motion scenarios.

Problem

• remote photoplethysmography
• heart rate estimation
• motion artifacts
• facial motion
• video-based measurement

Innovation

• angle-guided ROI optimization
• graph signal denoising
• remote photoplethysmography
• motion artifact suppression
• multi-region joint modeling