Scientists Uncover Invisible Motion in Video

A team at the Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory have developed a computer algorithm the detects movements that are so small or subtle as to be imperceptible to human eyes. The program then amplifies these movements kinetically or by color changes to the point that we can see them in video recordings. The program is called  Eulerian Video Magnification.

The team originally developed the program to monitor neonatal babies without making physical contact. But they quickly learned that the algorithm can be applied to other videos to reveal changes imperceptible to the naked eye. Prof. William T. Freeman, a leader on the team, imagines its use in search and rescue, so that rescuers could tell from a distance if someone trapped on a ledge, say, is still breathing.

“Once we amplify these small motions, there’s like a whole new world you can look at,” he said.

Some of the applications suggested for this new technology include medical diagnosis, finding oil, seeing how much a building sways, detecting lies, assessing manufacturing processes, and even determining whether a poker player is bluffing. See a video with more examples at the New York Times technology blog, Bits. Link -via Digg  

Newest 5
Newest 5 Comments

*whisperingly* Wow... wow. Thank you so much for posting this. It makes me wonder if LSD really does make one hyper-aware of time. I remember holding "Therapeutic Putty" while past my peak. I was living a Salvador Dali painting, watching every strand of time pass between my fingers. And it looked just like this. I just sat there and accepted time. Time does not stop. It was a beautiful truth, and one that now I have more clarity because of your posting this.
Abusive comment hidden. (Show it anyway.)
Login to comment.

Email This Post to a Friend
"Scientists Uncover Invisible Motion in Video"

Separate multiple emails with a comma. Limit 5.


Success! Your email has been sent!

close window

This website uses cookies.

This website uses cookies to improve user experience. By using this website you consent to all cookies in accordance with our Privacy Policy.

I agree
Learn More