Empirical tests of 16 architectures on 153 subjects show camera rPPG signals contain no recoverable subject-specific pulse morphology, with all models exhibiting template collapse.
Non -contact, automated cardiac pulse measurements using video imaging and blind source separation
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
An end-to-end spatial-temporal transformer with illumination augmentation, Residual Temporal Standardization Module, and hybrid waveform-spectral loss achieves 0.79 bpm MAE and 0.982 correlation on a new varied-illumination dataset, outperforming PhysFormer by 93.6% in MAE.
rPPG algorithms with Lp-norm and fractional-order peak enhancement achieve 1.92 bpm MAE for pulse rate and good HRV metrics vs ECG in 29 drivers, recommending 2SR for rate and CHROM for variability with 20 superpixels.
citing papers explorer
-
A Signal Extraction Approach for Remote Heart Rate Variability Assessment Using Proxy Measure in a Driving Simulator
rPPG algorithms with Lp-norm and fractional-order peak enhancement achieve 1.92 bpm MAE for pulse rate and good HRV metrics vs ECG in 29 drivers, recommending 2SR for rate and CHROM for variability with 20 superpixels.