‘Deepfake’ technology can now create real-looking human faces

A new study from Nvidia researchers show just how far artificial image-generation technology has come in recent years.

‘Deepfake’ technology can now create real-looking human faces
Karros et al.
  • In 2014, researchers introduced a novel approach to generating artificial images through something called a generative adversarial network.
  • Nvidia researchers combined that approach with something called style transfer to create AI-generated images of human faces.
  • This year, the Department of Defense said it had been developing tools designed to detect so-called 'deepfake' videos.

A new paper from researchers at Nvidia shows just how far AI image generation technology has come in the past few years. The results are pretty startling.

Take the image below. Can you tell which faces are real?

Karros et al.

Actually, all of the above images are fake, and they were produced by what the researchers call a style-based generator, which is a modified version of the conventional technology that's used to automatically generate images. To sum up quickly:

In 2014, a researcher named Ian Goodfellow and his colleagues wrote a paper outlining a new machine learning concept called generative adversarial networks. The idea, in simplified terms, involves pitting two neural networks against each other. One acts as a generator that looks at, say, pictures of dogs and then does its best to create an image of what it thinks a dog looks like. The other network acts as a discriminator that tries to tell fake images from real ones.

At first, the generator might produce some images that don't look like dogs, so the discriminator shoots them down. But the generator now knows a bit about where it went wrong, so the next image it creates is slightly better. This process continues until, in theory, the generator creates a good image of a dog.

What the Nvidia researchers did was add to their generative adversarial network some principles of style transfer, a technique that involves recomposing one image in the style of another. In style transfer, neural networks look at multiple levels of an image in order to discriminate between the content of the picture and its style, e.g. the smoothness of lines, thickness of brush stroke, etc.

Here are a couple examples of style transfer.

In the Nvidia study, the researchers were able to combine two real images of human faces to generate a composite of the two. This artificially generated composite had the pose, hair style, and general face shape of the source image (top row), while it had the hair and eye colors, and finer facial features, of the destination image (left-hand column).

The results are surprisingly realistic, for the most part.

Karros et al.

​Concerns over 'deepfake' technology

The ability to generate realistic artificial images, often called deepfakes when images are meant to look like recognizable people, has raised concern in recent years. After all, it's not hard to imagine how this technology could allow someone to create a fake video of, say, a politician saying something abhorrent about a certain group. This could lead to a massive erosion of the public's willingness to believe anything that's reported in the media. (As if concerns about 'fake news' weren't enough.)

To keep up with deepfake technology, the Department of Defense has been developing tools designed to detect deepfake videos.

"This is an effort to try to get ahead of something," said Florida senator Marco Rubio in July. "The capability to do all of this is real. It exists now. The willingness exists now. All that is missing is the execution. And we are not ready for it, not as a people, not as a political branch, not as a media, not as a country."

However, there might be a paradoxical problem with the government's effort.

"Theoretically, if you gave a [generative adversarial network] all the techniques we know to detect it, it could pass all of those techniques," David Gunning, the DARPA program manager in charge of the project, told MIT Technology Review. "We don't know if there's a limit. It's unclear."

How New York's largest hospital system is predicting COVID-19 spikes

Northwell Health is using insights from website traffic to forecast COVID-19 hospitalizations two weeks in the future.

Credit: Getty Images
Sponsored by Northwell Health
  • The machine-learning algorithm works by analyzing the online behavior of visitors to the Northwell Health website and comparing that data to future COVID-19 hospitalizations.
  • The tool, which uses anonymized data, has so far predicted hospitalizations with an accuracy rate of 80 percent.
  • Machine-learning tools are helping health-care professionals worldwide better constrain and treat COVID-19.
Keep reading Show less

Designer uses AI to bring 54 Roman emperors to life

It's hard to stop looking back and forth between these faces and the busts they came from.

Meet Emperors Augustus, left, and Maximinus Thrax, right

Credit: Daniel Voshart
Technology & Innovation
  • A quarantine project gone wild produces the possibly realistic faces of ancient Roman rulers.
  • A designer worked with a machine learning app to produce the images.
  • It's impossible to know if they're accurate, but they sure look plausible.
Keep reading Show less

Dark matter axions possibly found near Magnificent 7 neutron stars

A new study proposes mysterious axions may be found in X-rays coming from a cluster of neutron stars.

A rendering of the XMM-Newton (X-ray multi-mirror mission) space telescope.

Credit: D. Ducros; ESA/XMM-Newton, CC BY-SA 3.0 IGO
Surprising Science
  • A study led by Berkeley Lab suggests axions may be present near neutron stars known as the Magnificent Seven.
  • The axions, theorized fundamental particles, could be found in the high-energy X-rays emitted from the stars.
  • Axions have yet to be observed directly and may be responsible for the elusive dark matter.
  • Keep reading Show less

    Put on a happy face? “Deep acting” associated with improved work life

    New research suggests you can't fake your emotional state to improve your work life — you have to feel it.

    Credit: Columbia Pictures
    Personal Growth
  • Deep acting is the work strategy of regulating your emotions to match a desired state.
  • New research suggests that deep acting reduces fatigue, improves trust, and advances goal progress over other regulation strategies.
  • Further research suggests learning to attune our emotions for deep acting is a beneficial work-life strategy.
  • Keep reading Show less
    Surprising Science

    World's oldest work of art found in a hidden Indonesian valley

    Archaeologists discover a cave painting of a wild pig that is now the world's oldest dated work of representational art.

    Scroll down to load more…
    Quantcast