- A study has found that a computer can identify you based on your dance movements
- The research initially planned to study if machine learning can be used to identify which genre of music the participants were dancing to
- The computer was not able to give accurate results but was able to identify who among the 73 participants was dancing
Did you know that just like your fingerprints, your dance movements are also unique?
This information was accidentally found by a group of researchers at the University of Jyväskylä’s Centre for Interdisciplinary Music Research in Finland.
Science Daily reported that the group initially aimed to see of machine learning can be used to identify the genre of music (Blues, Country, Dance/Electronica, Jazz, Metal, Pop, Reggae, and Rap) the 73 participants were dancing to based on their movements. They were asked to listen to the music and “move freely to the randomized musical stimula, as they might in a dance club or party setting.”
Using machine learning, the authors looked into the movements of the participants based on the genres but their system got only 30 percent correct.
On the other hand, they found that 94 percent of the time, the computer figured out who among the 73 participants was dancing. “It seems as though a person’s dance movements are a kind of fingerprint. Each person has a unique movement signature that stays the same no matter what kind of music is playing,” data analyst and the research’s co-author Dr. Pasi Saari said.
It was also observed that the computer gave less accurate results when people were dancing to Metal music because “there is a strong cultural association between Metal and certain types of movement, like head banging. It’s probable that Metal caused more dancers to move in similar ways, making it harder to tell them apart,” according to Emily Carlson.
With this finding, Carlson said they hope to find the answers to more questions like whether a person’s movement signature will stay the same through his life or if cultural difference can be told based on movement signatures.
“Most research raises more questions than answers and this study is no exception,” she said.