Technology

Facebook Testing AI That Can Manipulate And Retouch Users’ Photos In A Very Convincing Way

No featured image available

Facebook has begun to experiment using an artificial intelligence program that can alter its users photos, the company announced Monday.

The “Eye In-Painting” uses General Adversarial Network (GAN), software that tricks users into thinking pictures are real, to draw in eyes when they’re closed in a picture. The system learns to recognize faces and paint in the eyes when they’re closed. Adobe has a similar “context aware fill” tool in Photoshop.

“Exemplar GANs provide a useful solution for image generation or in-painting, when a region of that image has some sort of identifying feature,” the research says. “They provide superior perceptual results because they incorporate identifying information stored in reference images or perceptual codes.”

The research further states that exemplar GANs “can be extended to other tasks within computer vision, and even to other domains.”

Facebook has other uses for AI on its platform. Co-founder and CEO Mark Zuckerberg said in testimonies to the U.S. congressional hearing and a hearing with the European Parliament that the company will be using software to automatically detect fake news and flag so-called hate speech.

“Some problems lend themselves more easily to AI solutions than others,” Zuckerberg said during his April 10 congressional testimony. “So, hate speech is one of the hardest because determining if something is hate speech is very linguistically nuanced. You need to understand what is a slur and what – whether something is hateful, not just in English, but the majority of people on Facebook use it in languages that are different across the world.”

Follow Kyle on Twitter @KylePerisic

Send tips to kyle@dailycallernewsfoundation.org

All content created by the Daily Caller News Foundation, an independent and nonpartisan newswire service, is available without charge to any legitimate news publisher that can provide a large audience. All republished articles must include our logo, our reporter’s byline and their DCNF affiliation. For any questions about our guidelines or partnering with us, please contact [email protected].