- Want lasting love? Learn to embrace the re-runs
- Three icons that are prime for deletion in 2020
- Mike Pence believes smoking isn’t dangerous. He will steer America’s coronavirus response
- Mass shooting shocks Milwaukee, as many as seven dead
- My therapist marginalised my climate anxiety, so I quit therapy
For whatever reason, this AI can identify you solely through your lame dance moves.
Your dance moves are so unique apparently, that AI can identify you by your unique dancing “fingerprint” with over 90% accuracy, according to a new study out of the Journal of New Music Research.
Researchers at Finland’s University of Jyväskylä have been utilising motion capture technology to study how one’s mood, personality and even ability to empathise is projected through their dance moves. While testing to see if AI could identify the music playing based on the participants’ style of dancing (it flopped, registering the incorrect genre over 70 percent of the time), they inadvertently found that the computer was able to correctly identify who was dancing over 94 percent of the time, regardless of what kind of music was playing.
Ok, so genre matters a little – the AI had a harder time identifying a dancer if said dancer was moving to Metal or Jazz music. Now, they’re not the most intuitive genres to groove to, and so us humans tend to fall back on generally the same types of movements. Movement of the head, shoulders and knees were the most vital markers the AI used when determining who a dancer was.
This explains why Metal would have made the task difficult for the computer; headbanging is a culturally engrained behaviour when it comes to “dancing” to Metal.
“Dance-recognition” software could go the way of face-recognition software eventually, but its current use is limited. At this point in time, researchers’ primary interest in the uses of this technology is not in surveillance but rather what the results show about how humans respond to music.
“We have a lot of new questions to ask,” says Emily Carlson, the first author of the study, “like whether our movement signatures stay the same across our lifespan, whether we can detect differences between cultures based on these movement signatures, and how well humans are able to recognize individuals from their dance movements compared to computers.”
There isn’t a reason to fret over being identified by your dance moves in a nightclub…yet.