Imagine that you could just point your iPhone’s camera at your baby and it would immediately tell you his vital signs: heartbeat and so on. Or that you could fire up an app and it could pick out tiny, invisible movements from what looks like a still video. Using a process called Eulerian Video Magnification, boffins at MIT are doing this already.
Eulerian Video Magnification takes a standard video feed and processes it in real time, applying something called Spatial Decomposition and then filtering the resulting components to identify minute differences between frames. These are then amplified, often to rather creepy effect.
You know how a colored filter applied to a B&W photo can cause a person’s complexion to go from milky to almost acne-ridden depending on the filter’s color? This is a little like that, only way more complicated.
The results allow you to see the blood flowing in and out of a person’s face as their heart pumps it around their body, for example, or to see the rhythmic movements of the pulse in the wrist. This has obvious medical applications, turning future iPhones into Star Trek-like Tricorders, but could also be used to detect subconscious thoughts through changes in expression.
But what it really is is neat. I love that you can take a relatively crappy video clip and immediately detect all kinds of hidden information within.