FCUS-rPPG: A Fast-Converging Unsupervised Framework for Remote Photoplethysmography via Gradient Oscillation Suppression

2026-06-01

AI Summary

This work addresses the challenges of unsupervised remote photoplethysmography (rPPG), which often suffers from slow convergence and limited generalization due to high gradient noise and unstable optimization. To overcome these issues, the authors propose the FCUS-rPPG framework, which jointly optimizes gradients, loss landscapes, and feature representations through spectral covariant feature disentanglement, a post-verification masking mechanism, perturbation-smoothed loss landscapes, and noise-aware null-space regularization. The method achieves rapid convergence with only a single training epoch and attains state-of-the-art performance across five cross-domain datasets, significantly outperforming existing unsupervised approaches that typically require tens to hundreds of training epochs.

Abstract

Remote photoplethysmography (rPPG) enables non-contact extraction of blood volume pulse (BVP) signals using consumer-grade cameras. Recent unsupervised rPPG methods learn BVP representations without requiring ground-truth physiological annotations, yet their optimization is often hindered by noisy and unstable gradients, resulting in slow convergence and limited cross-domain generalization. In this paper, we propose FCUS-rPPG, a fast-converging unsupervised rPPG framework with strong generalization capability. Motivated by the observation that BVP representations exhibit both multi-spectral covariation and low-dimensional manifold structure, we design a spectrally shared backbone that facilitates BVP feature disentanglement while improving optimization efficiency. To jointly enhance convergence stability and generalization performance, we further develop a unified optimization framework operating at the gradient, loss-landscape, and feature-representation levels. Specifically, a post-verification masking mechanism filters out misleading gradients according to the weak-amplitude physiological prior of BVP signals; a perturbation-based loss landscape smoothing strategy steers optimization toward more generalizable flat minima; and a noise-aware null-space regularization constrains feature updates to the orthogonal complement of the noise subspace, thereby mitigating noise-induced representation drift. Extensive experiments on five datasets demonstrate that FCUS-rPPG requires only one training epoch, whereas existing methods typically require tens to hundreds of epochs. Notably, FCUS-rPPG consistently achieves state-of-the-art (SOTA) performance in cross-dataset evaluations. This study provides an efficient and robust solution to the real-world deployment of unsupervised rPPG. The source code will be publicly available at https://github.com/JiaJieLee/FCUS-rPPG.

Problem

• remote photoplethysmography
• unsupervised learning
• gradient instability
• slow convergence
• cross-domain generalization

Innovation

• gradient oscillation suppression
• unsupervised rPPG
• fast convergence
• cross-domain generalization
• noise-aware regularization