[ad_1]
The false news is a phenomenon difficult to distinguish from the real news, not only because of the content of the articles and their viral effects on social networks, but also because it is easier to create now fake Realistic and attractive videos and photos. In addition, false news is a very detrimental practice as it influences people's behavior, election results and scams. Imagine what would happen if artificial intelligence was able to create fake but realistic video clips of people from a single photo of them. Samsung has just developed a new AI capable of doing this.
Inside an artificial intelligence lab in Russia, Samsung's artificial intelligence was able to make a fake video clip of people from a single photo or even a painting. And that seems pretty realistic. One of the video clips released shows Leonardo Da Vinci's Mona Lisa talking and moving. This table is the only image of the Mona Lisa used for the realization of the video, because it is the only existing one! Until now, creating deepfake videos required a lot of images to give a realistic result, but thanks to Samsung, only one photo is enough.
Samsung's new creation is called "realistic speaking neuron heads" and the word neural is not simply a stylistic choice. Neural networks are a type of machine learning capable of reproducing the human brain. They share a common point with Samsung's new algorithm. In order to create the video clip, the AI is watching a lot of videos to learn how human faces move, then applies everything that she has learned on a single photo.
Many sectors could benefit from such technology. , but we hope this is not a bad use, like creating fake viral videos.
TOP NEXT: Bicycle Pump Review Xiaomi Mijia: The First Portable Pump
Source link
