Welcome to the Qinsun Instruments Co., LTD! Set to the home page | Collect this site
The service hotline


Related Articles

Product Photo

Contact Us

Qinsun Instruments Co., LTD!
Address:NO.258 Banting Road., Jiuting Town, Songjiang District, Shanghai

Your location: Home > Related Articles > AI Defeats Humans Again: Distinguishing Personal Personality Based solely on a Self portrait

AI Defeats Humans Again: Distinguishing Personal Personality Based solely on a Self portrait

Author:QINSUN Released in:2024-01 Click:108

A new study suggests that artificial intelligence (AI) can once again defeat humans in identifying individual personalities based solely on a selfie. Compared to humans, machines can better recognize a feature and determine different personality traits of a species by analyzing the faces of characters. It is reported that a Russian research team has developed a new AI that is more advantageous than human and other competitive solutions.

The hierarchical architecture of Computer Vision Neural Network (NNCV) and Personality Diagnosis Neural Network (NNPD).

It is reported that this study aims to analyze the five personality traits of the target character (extroverted, neurotic, helpful, sincere, and responsible). The study involved approximately 12000 volunteers and 31000 selfies. In addition to providing selfies, volunteers were also asked to fill out a personality survey questionnaire.

It was found that when analyzing different selfies of the same person, AI would provide consistent personality judgments. In addition, compared to males, AI is more accurate in evaluating female selfies. In the end, the evaluation accuracy of the AI was "above average" and its overall performance was better than that of human volunteers.

Researchers explain that this AI machine has many potential uses, such as in the advertising industry to select models that better match product characteristics and are more popular among consumers.

The details of this study have been published in the recently published journal Scientific Reports.