A new paper published by Disney Research in partnership with ETH Zurich details a fully automated method for face swapping that uses a neural network to create incredibly high-definition likenesses.
It is the first example of face-swapping technology, also known as deepfake, that results in high-megapixel resolution, according to the researchers.
A unique approach
Recent Disney movies, such as Rogue One and Star Wars: The Rise of Skywalker have used face-swapping technology to have living actors perform the roles of other actors who have passed away with varying degrees of success.
The new method for face swapping, outlined by Disney, is so eerily accurate that it could be used in movies and TV and greatly improve the famous company's ability to bring actors back from the dead in a realistic way.
Aside from bringing actors back to life, the technology can be used for de-aging a living actor's face, making them look older, or also for replacing a stunt double's face with the face of the actor, the researchers say.
The new method is unique from other approaches in that any face used in a set can be swapped with any recorded performance. The technology is also able to recreate contrast and light conditions so that the actor looks like they were actually present in the same conditions as the scene.
The ethics of face-swapping tech
With Disney approaching such photorealism, videos could be produced by bad actors depicting 'real-life' events that never actually took place. It could also eventually mean that real video evidence could be brought into question the same way people shout 'fake news' today.
Thankfully, Disney has taken the academic route and has open-sourced its research so that others can determine methods for flagging and fighting malicious use of the technology.