In news programs, the simplest technologies are used to hide the identity of some people - image pixelation and voice distortion. This allows you to convey the essence of what the subject says, but not his personal emotions regarding the subject of the conversation. And they are sometimes no less important, so at the University of British Columbia they taught artificial intelligence to redraw people's faces beyond recognition, but at the same time preserve their emotions.
The work begins with the fact that the operator manually corrects the original image, removing obvious marker details that could indicate a person's personality. He also places tags that describe the subject's emotions. The AI then makes a quick and easy randomization of the pixel composition so that the original image cannot be accurately reconstructed by reverse editing.
Then real creativity begins, where AI copies the technique of Van Gogh, Picasso or other famous painters in order to draw new facial features. He is looking for a form that can combine the inner state of a person and his outer manifestation. At the same time, accuracy and realism are deliberately ignored, the proportions, shades and position of individual objects begin to take on grotesque features. You no longer recognize in the picture the one from whom you drew it, but you will understand that the person smiles sadly or is ready to scream with fear.
Frame by frame, AI can redraw any video, creating groundbreaking new content. Not only for the needs of the media, but also, probably, laying the foundation for a new direction of the visual arts. Imagine a film of intimate stories from people who would never have revealed them if they were not sure that no one would personally recognize them, but on the other hand, viewers will understand the full range of experiences and the emotional background of those events. And the technology is also suitable for use in virtual reality systems.