In the Fox TV show "Lie to Me," Dr. Cal Lightman was able to tell whether someone was lying by observing what he called "micro expressions" on their faces. The twitch of an eye, the quickening of a pulse, the beads of sweat on a brow -- he looked for clues too subtle for most of us to catch.
Now, researchers out of MIT are developing a video technology they call Eulerian Video Magnification that could do that and more -- by amplifying the motion in a standard video sequence to detect information not visible to the naked eye.
"Our goal is to reveal temporal variations in videos that are difficult or impossible to see with the naked eye and display them in an indicative manner," the researchers write. "Our method... takes a standard video sequence as input and applies spatial decomposition, followed by temporal filtering to the frames. The resulting signal is then amplified to reveal hidden information."
They've already been able to amplify images to the point of being able to reveal the flow of blood pulsing through the human face by detecting subtle differences in color, as well as to amplify small motions such as an infant's chest rising and falling.
The potential for medical sensors and monitors that require only video, instead of body contact, is immense. We've seen this sort of approach before (also out of MIT), but the applications could end up being quite broad for even the most casual user.
Imagine, for instance, taking a quick video of a first date's face with your cell phone to detect arousal, or the absence thereof. Time was, we had to rely on instinct for this kind of detection. Soon, for better or worse, we may have certainty instead.