Deepfake Technology, a pioneer in the United Kingdom, is the director of a company in the United Kingdom, who believes that if we only have three years of age, to have computer versions of so well-known actors to distinguish them from real human beings.
Victor Riparbelli, 27, founded two years ago by Synthesia. When the company made its first reception in 2018, it seems that a new BBC anchor technology used to speak Spanish, Mandarin and Hindi.
Recently, the company applied its technology to David Beckham football soccer. In collaboration with the Malaria Must Die campaign, he manipulated the functions of Synthesia Beckham's facial in nine languages where nine malaria survived.
You can see this video here:
Riparbelli told Business Insider that recording was almost the same day newspaper. The only difference is that before Beckham delivered his lines, he had to train the synthesis algorithm on his face.
To do this, Beckham had to talk to the camera. There's no need for scripting, even though Riparbelli helps keep quizzes in mind about what people are saying. An example, they said what they would say about people about breakfast.
This footage is translated to the training data, the algorithm teaches how to move Beckham's face to create a digital model. The entire process lasts about three to four minutes. "Once this is done, it does not require any particular hardware or camera, or something like that," Riparbelli said.
You can watch more videos here: Video by Synthesia:
It creates a horrible rock
Synthesia announced this week that it has received $ 3.1 million, some of which were Mark Cuban's first investors. Riparbelli told Business Insider that the company was focused on advertising, but it is intended to incorporate the television and film into the world of special effects.
"As the company progresses, we will expand our platform and work with cinema and entertainment, and we will have the opportunity to live a lot of ideas. [more easily] More than today, "he said. He has said he is developing a synthesis technology, which is the same process that is already used in Hollywood films," we are doing it with a totally automatic network of neurons. "
Read more: Scarlett Johansson says "trying to make people into deep pornographic videos is a" lost cause "
In recent years, we have seen more and more CGI actors and actors in movies: think of "Bladerunner" or "Star Wars: Rogue One." However, many of these digital actors are known as the "amazing valley": very realistic, beautiful, not realistic, to be completely convincing. There is little.
Riparbelli thinks we're very close to getting rid of the furtive valley.
"In the next three years we will see a significant improvement in how digital humans are created," he said. Synthesia has said he can do a photorealistic man, "we can still do it with movies."
Business Insider has asked if the company is interested in applying its technology to video games. "Our focus is nowhere, our vision is to produce a photorealistic video, but who knows one day in the future," he said.
Synthesis "can not" escape from deep failures "
Synthesia's success has reflected deep insights, the AI software uses realistic and often disturbing ways of mapping people's faces into the body of others.
Technology has won a great deal, since they have used face-to-face pornography on pornographic pornography, and has caused more general fears about misinformation and false news. Inside Synthesia prefers to use the word "synthesis" to escape from negative connotations.
However, Riparbelli says that the outside world will continue to describe them deeply. "We can not run away, and we're fine."
"Would we not always connect with that term? Yes, but in a different way everyone knows what you are talking about" when you talk about deep-skinned words, there's really no other word. "
Riparbelli does not find the profound world to be completely annoying. He has said that his favorite is a deep video embedded in the face of Steve Buscemi's Jennifer Lawrence.