In the not-so-distant future of hyper-surveillance, could your "dad dance" be used to identify you in the middle of a bustling club?
New research by the University of Jyväskylä in Finland has used motion capture technology and machine learning to understand how different people shimmy and groove to music. It turns out, your dance moves are almost as unique as your fingerprint and can be used to personally identify you with a surprising degree of accuracy. Furthermore, the team also says that the way you dance could be used to understand your personality and gain insights into the mood you're in.
"It seems as though a person's dance movements are a kind of fingerprint," Dr Pasi Saari, co-author of the study and data analyst, said in a statement. "Each person has a unique movement signature that stays the same no matter what kind of music is playing."
As reported in the Journal of New Music Research, a total of 73 volunteers were played eight genres of Western music – these included Blues, Country, Dance/Electronica, Jazz, Metal, Pop, Reggae, and Rap – and simply told to dance in any way that felt natural. The motion-tracking technology kept tabs on dozens of points on the body, watching the direction and speed at which they moved.
The original idea behind the study was to see if computers can guess which genre of music participants were dancing to by analyzing their movements, but it proved to be surprisingly bad at identifying the genre. The “easiest” to guess was Metal music, which was accurately identified around 53 percent of the time, but Pop music was identified less than 10 percent of the time.
It did, however, appear to be remarkably accurate at guessing which of the participants was dancing.
While the computer was only able to correctly identify the music genre around 30 percent of the time, it managed to correctly identify which of the 73 people was dancing around 94 percent of the time. Regardless of the genre, the computer was able to tell.
A similar piece of research, published in 2017, found that a person’s walk can also provide insights into their personality and mood.
Fortunately, the researchers are confident that this field of research won’t be used for sinister ends, such as mass-surveillance, just yet. Instead, they hope to build up more evidence about what our movements say about us, our personality, and our culture.
“We’re less interested in applications like surveillance than in what these results tell us about human musicality,” explained Dr Emily Carlson, the first author of the study. “We have a lot of new questions to ask, 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.
“Most research raises more questions than answers and this study is no exception.”
57