There is a difference in the way that men and women smile. Researchers from the University of Bradford have created an artificial intelligence (AI) system that can figure out your sex based only on your smile.
While there are automatic sex recognition systems available, those systems can only use static images in comparing fixed facial features. But the new system is the first to use the dynamic movement of a smile to distinguish if the user is a man or a woman.
In order to develop this new tech, the team mapped 49 landmarks on the human face. Most of the landmarks were around the eyes, mouth and down the nose. Then they used these landmarks to assess how a human face changes when a person smiles using underlying muscle movements. These small muscle movements include changes in distances between the points and how a smile flows. Then they tested the noticeable differences between a male and a female smile. One of the major differences they found was that women’s smiles are typically more expansive.
Professor Hassan Ugail, the lead researcher, said “Anecdotally, women are thought to be more expressive in how they smile, and our research has borne this out. Women definitely have broader smiles, expanding their mouth and lip area far more than men."
An algorithm was then created using their analysis. They tested it against video footage of 109 people smiling. The computer was 86 percent accurate in determining the sex of the person smiling. The team aims to improve this accuracy.
"We used a fairly simple machine classification for this research as we were just testing the concept, but more sophisticated AI would improve the recognition rates," said Professor Ugail.
One of the main reasons that this technology was developed was to enhance machine learning capabilities. But since its introduction, many people have asked questions that have inspired the research team to investigate and develop this system further.
One of these questions is how the system may react to a transgender person’s smile. They also wonder how plastic surgery will affect the system’s recognition.
"Because this system measures the underlying muscle movement of the face during a smile, we believe these dynamics will remain the same even if external physical features change, following surgery for example," said Professor Ugail. "This kind of facial recognition could become a next-generation biometric, as it's not dependent on one feature, but on a dynamic, that's unique to an individual and would be very difficult to mimic or alter."
The paper on this research was published in The Visual Computer: International Journal of Computer Graphics.