Photofeeler has created a neural network that identifies the person’s attractiveness in photography.

According to the authors of the project, their development is trained on hundreds of millions of people’s thoughts, so it works better than analogues. The algorithm evaluates the appearance of a person, exposing the assessment of three criteria: attractiveness, trust and intelligence.

Neural network operates both with a photo, and with a real-time recording. Developers recommend using this algorithm to select the best photos and accessories.

The company has just started to test artificial intelligence and launched it on the site by rating the photo. True, there the algorithm is still used only as a curator.