Combining Magnification and Measurement for Non-Contact Cardiac Monitoring
Abstract
Deep learning approaches currently achieve the state-of-the-art results on camera-based vital signs measurement. One of the main challenges with using neural models for these applications is the lack of sufficiently large and diverse datasets. Limited data increases the chances of overfitting models to the available data which in turn can harm generalization. In this paper, we show that the generalizability of imaging photoplethysmography models can be improved by augmenting the training set with "magnified" videos. These augmentations are specifically designed to reveal useful features for recovering the photoplethysmogram. We show that using augmentations of this form is more effective at improving model robustness than other commonly used data augmentation approaches. We show better within-dataset and especially cross-dataset performance with our proposed data augmentation approach on three publicly available datasets.
Cite
Text
Nowara et al. "Combining Magnification and Measurement for Non-Contact Cardiac Monitoring." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021. doi:10.1109/CVPRW53098.2021.00422Markdown
[Nowara et al. "Combining Magnification and Measurement for Non-Contact Cardiac Monitoring." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021.](https://mlanthology.org/cvprw/2021/nowara2021cvprw-combining/) doi:10.1109/CVPRW53098.2021.00422BibTeX
@inproceedings{nowara2021cvprw-combining,
title = {{Combining Magnification and Measurement for Non-Contact Cardiac Monitoring}},
author = {Nowara, Ewa Magdalena and McDuff, Daniel and Veeraraghavan, Ashok},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2021},
pages = {3810-3819},
doi = {10.1109/CVPRW53098.2021.00422},
url = {https://mlanthology.org/cvprw/2021/nowara2021cvprw-combining/}
}