Heart Rate Estimation from Face Videos

Description:

Using a self-adaptive matrix heart rate etimation from face videos is improved.

 

 Background: 

Recent studies in computer vision have shown that, while practically invisible to a human observer, skin color changes due to blood flow can be captured on face videos and be used to estimate the heart rate (HR). These subtle changes encompass both color and motion, and they are induced by the internal functioning of the heart. However, state-of-the-art approaches are not robust enough to operate in natural conditions (e.g. in case of spontaneous movements, facial expressions or illumination changes). Moreover, conventional methods require specific and invasive  contact sensors to be attached to the human skin and/or a static location for non-contact remote HR measurement.

 Technology Overview:  

The present invention addresses the problems of HR estimation from face videos in realistic conditions. To cope with large facial variations due to spontaneous facial expressions and movements, a principled framework is provided to automatically discard the face regions corresponding to noisy features and only use the reliable ones for HR prediction. The region selection is addressed within a novel matrix completion-based optimization framework, called self-adaptive matrix completion, for which an efficient solver is proposed. The approach is demonstrated to be more accurate than previous methods for average HR estimation on publicly available benchmarks. In addition, short-term analysis results show the ability of the present method to detect instantaneous heart rate.

 

https://binghamton.technologypublisher.com/files/sites/pexels-photo-4157791.jpeg

https://www.pexels.com/photo/blur-chart-check-up-curve-415779/

 

 

 Advantages:  

  • HR measurement while simultaneously selecting the most reliable face regions for robust HR estimation.
  • Higher accuracy while operating on short time sequences to detect instantaneous HR.
  • Non-static (i.e. subjects to be monitored need not to be still

 Intellectual Property Summary: 

 

U.S. 10,335,045

 

Binghamton University RB502

 

Patent Information:
For Information, Contact:
Scott Hancock
Senior Director, Technology Transfer
Binghamton University
(607) 777-5874
shancock@binghamton.edu
Inventors:
Lijun Yin
Keywords:
#SUNYresearch
Technologies
© 2024. All Rights Reserved. Powered by Inteum