Our emotions can affect everything, from appetite to the way we perceive the world – and even the way we walk.

So can we explain someone’s feelings just based on their gait? That’s exactly what scientists at the University of North Carolina at Chapel Hill and the University of Maryland at College Park have been teaching computers to do. Using deep learning, their software can analyze a video of someone walking, convert it into a 3D model and extract that person’s gait. A neural network will then determine the dominant motion and pair it with a specific type of emotion, based on the data it has been trained before. According to the researchers, this deep learning model can predict 4 different emotions – happy, sad, angry and normal – with an accuracy rate of up to 80%.

This deep learning model can predict 4 different emotions – happy, sad, angry and normal.

Although we’ve seen many AI trained to guess human emotions based on facial expressions or voices, this is the first time there’s an algorithm trained to correctly guess emotions just by looking at them. people walk.

Aniket Bera, a research supervisor and a professor of computer science at the University of Maryland, said their research was not about seeking ways to detect actual emotions, but rather predicting perceived emotions. – that is, one’s emotions in the eyes of the opposite person, just like we predict each person’s emotions when they meet each other every day. According to Bera, through research, it is possible to teach robots how to guess what the people around them are thinking, and change their behavior accordingly. Or vice versa: it can help engineers design robots that better communicate with their gait and body movement.

Bera added that this research will help pave the way for future surveillance applications, or even help make the mixed reality experience more compelling, as human 3D models are following suit. Specific ways can help design more realistic characters.

But the team concludes that the next step will be to focus on other actions, not just walking, such as running, and gestures, to understand even the smallest emotions we express. reached on the move.