algorithm increased to 91% and 83%, respectively, given five facial images per person . of facial images of 40 gay and 40 heterosexual men, and found that on .
We used deep neural networks to extract features from 35,326 facial images. Prediction models aimed at gender alone allowed for detecting gay males with.
When the algorithm was shown five facial images per person in the pair, to distinguish between gay and straight people,” Kosinski tells me.
Artificial intelligence can accurately guess whether people are gay or With billions of facial images of people stored on social media sites and.
The algorithm took into account both 'fixed' and 'transient' facial features sexual orientation from facial images'/Yilun Wang, Michal Kosinski).