Anonymous blurred faces on the news given more emotion thanks to AI
Image credit: simon fraser uni
A new AI process has been developed to blur faces in order to maintain anonymity in news reports while maintaining more of the emotional expression than the traditional pixelated approach.
A research team from SFU’s School of Interactive Arts and Technology (SIAT) created an updated pixelating technique using an AI “painting” approach to anonymisation.
“Our goal is to create a working technique that would be much better at conveying emotional and knowledge information than current anonymisation techniques,” said Professor Steve DiPaola who led the team.
“When artists paint a portrait, they try to convey the subject’s outer and inner resemblance," he said. “With our AI, which learns from more than 1,000 years of artistic technique, we have taught the system to lower the outer resemblance and keep as high as possible the subject’s inner resemblance - in other words, what they are conveying and how they are feeling.
“Our system uses five levels of AI processing to simulate a smart painter, using art abstraction to repaint the video as if an artist was painting every frame. The result is more engaging, especially since not everyone listens to stories - so the art component becomes more relevant.”
The system doesn’t change the pixels of the video frames as Adobe-like systems can, but instead produces a painting-style result of every frame.
“It’s actually an open and dynamic process that allows levels of control throughout. We eventually want to the subject or producer to be able to customise the final result based on their needs,” DiPaola said.
The tool’s effectiveness at anonymising while retaining a strong degree of emotional connection or resonance should result in better final product for anonymised video, especially in 360 or VR. The team plans to put the technology to work in a variety of projects. The team has already attracted interest from several major media companies looking to explore new media approaches, including the New York Times, Washington Post and the BBC.
The technology is similar to ‘Vincent’, an AI project developed by Nvidia last year that can apply the style of famous artists to any picture fed through the system.