AI can recognise each person’s dancing ‘fingerprint’: Study

Each person’s dancing style has a unique signature - regardless of the type of music - and this pattern can be accurately identified by a computer, according to a study which may lead to a deeper understanding of how music affects humans.

The study, published in the Journal of New Music Research, used motion capture technology – the kind used in Hollywood – to learn what a person’s dance moves say about their mood, how extroverted or neurotic they are, and how much they empathize with others.

“We actually weren’t looking for this result, as we set out to study something completely different,” said Emily Carlson, the first author of the study from the University of Jyvaskyla in Finland.

“Our original idea was to see if we could use machine learning (ML) to identify which genre of music our participants were dancing to, based on their movements,” Carlson said.

In the study, 73 participants were motion captured as they danced to eight different genres Blues,

Country, Electronica, Jazz, Metal, Pop, Reggae and Rap.

The participants were instructed to listen to the music and move any way that felt natural.

“We think it’s important to study phenomena as they occur in the real world, which is why we employ a naturalistic research paradigm,” said Petri Toiviainen, senior author of the study.

Using ML, a form of artificial intelligence, the researchers tried to distinguish between the musical genres by just analysing the dancers’ movements.

The ML algorithm was able to identify the correct genre less than 30 per cent of the time.

However, system could correctly identify which of the 73 individuals was dancing 94 per cent of the time.

“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,” said Pasi Saari, another co-author of the study.

According to the researchers some genres may have more effect on individual dance movements than others.

They said the computer was less accurate in identifying the participants when they were dancing to Metal music.