Key result
A two-stage deep learning scheme for non-contact heart rate estimation from low-resolution facial videos achieved mean absolute percentage errors of 3.320 and 4.231 across two datasets.
Why the study?
Does a two-stage deep learning scheme improve heart rate estimation accuracy from low-resolution facial videos compared to high-resolution approaches?
Does a two-stage deep learning scheme improve heart rate estimation accuracy from low-resolution facial videos compared to high-resolution approaches?
A novel two-stage deep learning scheme effectively recovers rPPG information from low-resolution facial videos, achieving heart rate measurement accuracy comparable to state-of-the-art methods using high-resolution images.
Remote photoplethysmography (rPPG) enables non-contact heart rate (HR) estimation using facial videos and exhibits significant convenience over traditional contact-based HR measurement approaches. However, its effectiveness depends mainly on the number of pixels of the facial region; thus, the estimation accuracy is easily degraded due to missing rPPG information when the resolution of facial images is low. Given that in most real application scenarios, such as group-oriented and long-distance vital signs measurement, only low-resolution facial videos are available, finding a solution to recover rPPG information is a key to improve the performance of non-contact HR measurement. In this paper, we propose a two-stage deep learning scheme to achieve accurate HR measurement from low-resolution facial videos. In the first stage, an rPPG information recovery network is proposed to recover the rPPG information during video super-resolution (SR) processing. The up-scaled images are then fed into a temporally-aware HR measurement network in the second stage. An attention mechanism that reassigns the weights of the temporal information in each feature channel is designed to improve the measurement accuracy. The mean absolute percentage error achieves 3.320 and 4.231, while the Pearson’s correlation coefficient reaches 0.930 and 0.892 on two datasets respectively. Experimental results show that rPPG information could be effectively recovered through SR processing. Compared with the state-of-the-art approaches that treat high-resolution images as the input, the proposed method achieved closely comparable measurement performance using merely low-resolution images.
No takes yet. Share an insight, caveat, or question.
Yue et al. (2021) studied Heart rate estimation. Two-stage deep learning scheme for rPPG information recovery vs. State-of-the-art approaches using high-resolution images was evaluated on Mean absolute percentage error and Pearson's correlation coefficient. A two-stage deep learning scheme for non-contact heart rate estimation from low-resolution facial videos achieved mean absolute percentage errors of 3.320 and 4.231 across two datasets.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: