"our facial expressions vary greatly depending on the context, and our cultural background," explains Martinez. "And it is important to realize that not all when they smile and are happy and not all happy people are smiling. I would go even further: a lot of people when you do not smile, are not necessarily unhappy. And if you're happy for the whole span of a day does not go around the street with a smile perpetually in the face: you're going to be happy."
A simple smile and then does not tell us very much. In some countries, on the other hand smile often it is a good social norm. While others could make you pass for insane. Not surprisingly, when Martinez and his team have tried to analyze the movements of the facial muscles in human and put them in relation with the emotions of a group of volunteers, the results have proved to be almost always incorrect. We human beings try to read the mood and behavior of the next also through its expressions, it is true, but we base our reviews (although always fallible, bear in mind) on a series of clues, the most wide: the complexion of the face, posture, movements of arms and legs, and context information. For a computer things are different: more and more often we hear talk of artificial intelligences that use facial recognition to predict human intentions. For example, in the field of security, indicating the people who are going to commit a crime, or even to school, where it speaks of recognizing automatically the students who are not paying attention to the lesson.
A business in which you are launching a growing number of companies, with results that Martinez defines disappointing. Together with his team has indeed put to the test many of the technologies of this type available today, finding them mostly ineffective. According to the expert, the problem is inherent to the nature of these algorithms: use only the movements of the face is not enough to read the human behavior.
In one of his experiments showed, for example, a group of volunteers a lot of photos close-up of a human face: the mouth opened in a scream, a silent, tense muscles, complexion reddened. "When people looked at the photos thought: this person must be really pissed off or extremely angry, to the point of screaming," says Martinez. "But when they saw the whole picture you are faced with a player who was screaming for celebrating a goal".
in Short, judging people based solely on his facial expressions is not a good idea. Nor for a human being, much less for a machine. Martinez, however, is optimistic: sooner or later artificial intelligences will be powerful enough to take into consideration also other types of clues, social, postural, etc... to get to interpret the intentions of human beings. Always, however, taking into account two important facts that relate to the technology in general. "The first - concludes the expert - is that no technology will ever be 100% accurate. And the second is that in order to decipher the intentions of a person serves much more of a look at the expression of his face is something that will need to understand both the people who work in this field, algorithms from their products".
"The Republic will fight always in defense of the freedom of information, to its readers and to all those who have at heart the principles of democracy and civil coexistence"Carlo Verdelli SUBSCRIBERS TO REPUBLIC © Reproduction reserved Today about Flavio Bucci, what dream actors when they get older the Count goes on pressing on the Italia living: “in or out in a few days” Decrees yet to change. Steps forward, but the understanding there is A step too short-tear of Renzi: at the Regional Italy Viva will run by itself