Search papers, labs, and topics across Lattice.
The paper introduces FreqPhys, a novel rPPG framework that leverages physiological frequency priors to improve robustness against motion artifacts and illumination changes. It employs physiological bandpass filtering, spectrum modulation with adaptive spectral selection, and cross-domain representation learning to fuse spectral priors with time-domain features. A frequency-aware conditional diffusion process then reconstructs high-fidelity rPPG signals, demonstrating state-of-the-art performance on six benchmarks, especially under motion.
Explicitly modeling physiological frequency priors in rPPG unlocks significant improvements in signal recovery, even under challenging motion conditions that plague existing time-domain methods.
Remote photoplethysmography (rPPG) enables contactless physiological monitoring by capturing subtle skin-color variations from facial videos. However, most existing methods predominantly rely on time-domain modeling, making them vulnerable to motion artifacts and illumination fluctuations, where weak physiological clues are easily overwhelmed by noise. To address these challenges, we propose FreqPhys, a frequency-guided rPPG framework that explicitly leverages physiological frequency priors for robust signal recovery. Specifically, FreqPhys first applies a Physiological Bandpass Filtering module to suppress out-of-band interference, and then performs Physiological Spectrum Modulation together with adaptive spectral selection to emphasize pulse-related frequency components while suppress residual in-band noise. A Cross-domain Representation Learning module further fuses these spectral priors with deep time-domain features to capture informative spatial--temporal dependencies. Finally, a frequency-aware conditional diffusion process progressively reconstructs high-fidelity rPPG signals. Extensive experiments on six benchmarks demonstrate that FreqPhys yields significant improvements over state-of-the-art approaches, particularly under challenging motion conditions. It highlights the importance of explicitly modeling physiological frequency priors. The source code will be released.
