Disney’s New AI Software Can Easily Make Actors Look Younger or Older

Disney Research Studios has revealed a production-ready face re-aging system for visual effects that can make actors look younger or older easily, accurately, and quickly.

Spotted by Gizmodo, Disney’s new artificial intelligence-powered (AI) method solves many of the problems previous attempts have come up against. The method is able to generate convincing-looking results in high resolution and takes less than five seconds per frame to produce.

In its abstract, the researchers explain that current methods for aging and de-aging actors using facial imaging methods do exist, they offer little practical use. Some will successfully age or de-age a subject, but that end result bears little resemblance to the original person. They also tend to have poor resolution and the results are often unstable between frames, which makes them very unrealistic to a viewer.

The Disney Research Studios method is advertised as fixing all of these problems and is presented as the first practical, fully-automatic, and production-ready method for re-aging faces in video.

“Our first key insight is in addressing the problem of collecting longitudinal training data for learning to re-age faces over extended periods of time, a task that is nearly impossible to accomplish for a large number of real people. We show how such a longitudinal dataset can be constructed by leveraging the current state-of-the-art in facial re-aging that, although failing on real images, does provide photoreal re-aging results on synthetic faces,” the researchers explain.

“Our second key insight is then to leverage such synthetic data and formulate facial re-aging as a practical image-to-image translation task that can be performed by training a well-understood U-Net architecture, without the need for more complex network designs,” the researchers continue. “We demonstrate how the simple U-Net, surprisingly, allows us to advance the state of the art for re-aging real faces on video, with unprecedented temporal stability and preservation of facial identity across variable expressions, viewpoints, and lighting conditions.”

The team says that its face re-aging network, or FRAN, uses simple and intuitive mechanisms that give artists localized control and the freedom to fine-tune and direct the aging effect on a subject, control that the researchers says is extremely important in production environments and is typically overlooked in research work.

The team’s full paper on the new technology can be downloaded from the Disney Research Studios website.