In this article, we report on some data characteristics of hemodynamic oscillations obtained via remote photoplethysmography (rPPG). To implement this method we used camera Mako G-050C and a personal computer. The proposed algorithm for rhythmogram rPPG formation includes the following steps: 1. Detection of face position by Haar cascade classifiers 2. Extraction «region of interest» (ROI) from forehead 3. Exclusion areas with no exposed skin from ROI 4. Division of ROI into 6 partial sections 5. Formation of partial signals via calculation of average image brightness of green channel in each partial section 6. Linear interpolation of partial signals with sampling frequency 30 Hz 7. Normalization of partial signals amplitudes 8. Application of principal component algorithm (PCA) for amplification of rPPG signal 9. Filtration of amplified signals by bandpass filter for cutoff frequencies from 0,667 to 2,5 Hz 10. Formation of rhythmogram time series The presented results show that rPPG can be used for an estimation of slow hemodynamic oscillations in frequency range from 0,003 to 0,04 Hz with high reliability. These oscillations have a high level of correlation (above 0,9) with very low frequency (VLF) spectral component of heart rate variability (HRV). That allows one to use rPPG in medicine (to monitor psychological, emotional and energy insufficient state changes) and in security systems