New York University’s GreenDot Project attempts to computerize the identification of individuals on the basis of unique, unconscious movements and gestures as exclusive to each person as their fingerprints.
The goal of the project is to train a computer to recognize a person based on his or her motions, and to identify the person’s emotional state, cultural background, and other attributes.
In the demo, above, the computer trains its electronic eyes on footage of Democratic presidential nominee Barack Obama.
To identify who is in the video, the computer first looks for movement in the scene. Green dots indicate motion. As the video plays, and the computer collects motion data, it can eventually isolate and identify the human.
As for why this research is being done, the GreenDot site is elusive.
The current focus is the analysis of national and international public figures while they are giving speeches, with future plans to investigate many other domains.
“Other domains”?
NYU is in Lower Manhattan, 2 1/4 miles from Ground Zero. Use your imagination. Or use your illusion.
[via BoingBoing]
1 comment so far ↓
word! train a computer to distinguish between eminem’s rap gestures and that of a “real” mc. perhaps intelligent machines will not overthrow humanity but finally give black people objective means to gain control over their gestures. it reminds me of how the fremen of dune had to walk off-step to avoid detection by the sandworms.
“buck buck buck, take that witcha on the way down!”
– kool g. rap
Leave a Comment