Scientists reverse engineer 3D facial models using info ‘stored in brain’
The work, carried out by neuroscientists from the University of Glasgow, was published Monday in the journal Nature Human Behavior. Their findings will be the “cornerstone for greater understanding of the brain mechanisms of face identification,” according to the team, and could have a future in AI and gaming technology, as well as eyewitness testimony.
The team of 14 scientists initially tested the theory by determining what specific facial information they recalled to identify four other colleagues from memory. Then, researchers had volunteers compare faces which were either the same age, gender or ethnicity, but differed in terms of their facial identity.Also on rt.com Life imitates Black Mirror: 5 real-life technologies straight from the disturbing series
Eventually the team was able to “crack the code” of what defines visual identity and generate it with a computer program. Over many trials, the group finally devised a method to reconstruct the information taken from a person’s brain to recreate a specific identity from memory.
The main hurdle in this scientific breakthrough was understanding the information people store in their memory to recognise a specific face, “but we have developed a tool which has essentially given us a method to do just that,” said Philippe Schyns, Professor of Visual Cognition at the Institute of Neuroscience and Psychology.
Schyns explained that the team could graphically render the information by “reverse engineering the information that characterizes someone’s identity, and then mathematically representing it.”
If you like this story, share it with a friend!