Eulerian Video Magnification is a series of algorithms used to amplify subtle changes in a video (positional, colour) to show movements originally invisible to the human eye. I've always had a passion for health technology so I wanted to apply this algorithm to detect human heart rate from a simple video taken from a cell phone, completely contactless.
This project is currently a work in progress to better implement Eulerian Video Magnification, however, in its current state it can still detect heart rate from a video. The following graph below uses face.mp4 and plots the change in colour intensity.
This is currently the most accurate method from this project, a live implementation is also being implemented.
- Thanks to MIT CSAIL for an in-depth research paper on Eulerian Video Magnification
- Thanks to Steve Mould for the YouTube Video that inspired this project