Researchers have developed new computational tools that help computers determine whether faces fall under categories like attractive or threatening, according to a recent paper published in the journal PLoS ONE.
Mario Rojas and other researchers at the Computer Vision Center in the Autonomous University of Barcelona in Spain, along with researchers from Princeton University’s department of psychology, developed software that is able to predict these traits, in some cases with accuracies beyond 90 per cent, according to a release.
Facial characteristics play an important role in our everyday assessment of people. “The perception of dominance has been shown to be an important part of social roles at different stages of life, and plays a major role in mate selection,” said Rojas. If information on which the evaluation of faces is based could be automatically learned, it could be used as a tool for designing better interactive systems.
The team studied to what extent this information was learnable from the perspective of computer science. Specifically, the task was formulated with the intention of predicting nine facial trait judgments (attractive, competent, trustworthy, dominant, mean, frightening, extroverted, threatening, and likable) using machine-learning techniques (a branch of artificial intelligence that uses examples to teach a programme how to work).
The team trained and tested their algorithm on a set of synthetic facial images generated in a previous study. People were asked to describe and rate a set of facial images, and the results were used to generate synthetic facial images, each associated with specific traits such as trustworthiness or dominance. The researchers used a subset of these images, together with their labels, to ‘teach’ the computer to read a face, and tested the prediction accuracy using the rest of the images. Three traits, dominant, threatening and mean, were found to be predictable with accuracies between 91 per cent and 96 per cent. The study also aimed to find what information was computationally useful for the prediction.