Skip to main content

Forget painting-style transfers, this A.I. creates realistic portraits of fake people

A Style-Based Generator Architecture for Generative Adversarial Networks

After selling the first artificial intelligence generated artwork earlier this year, deep learning algorithms are now tackling portraits — of made-up people that don’t exist. In a research paper, Nvidia recently shared the results of a generative adversarial network (GAN) trained to generate images of people while unsupervised.

The concept is based in part on style transfer technology, which transfers the style of one image onto another in order to do anything from building a photo app like Prisma to creating realistic deep fakes. Nvidia researchers redesigned the GAN to use a style transfer, while adjusting that style on each layer of code. The change, the researchers say, allows the software to generate high-level attributes like subtle differences in poses as well as random variation in features.

The researchers mixed the adjusted algorithms with a wider training dataset, one of which is the faces on Flickr instead of a database of celebrity faces, and adjusted the way the generator embeds the code for more realistic results that aren’t simply re-creating an existing person in a new photo. The researchers also switched between two different randomized codes for more variation in features.

While the GAN can randomly create images of people, the researchers also managed to control the results by using a style transfer to mix qualities from two different portraits. A “style” image mixed with a “source” image uses features like skin, eye, and hair color from the style image and applies it to the source image, which keeps the original’s gender, age and pose. The lighting styles and background of the style image also transfer to the source image.

The results show dramatic improvement over similar Nvidia research from four years ago, where the resulting images were black and white images with few details. The program’s not perfect — the system, for example, created some portraits with two different colored eyes and in others didn’t maintain facial symmetry.

While the paper shows impressive progress, the researchers didn’t detail the real-world use of such algorithms. Like the technology used for deep fakes, the ability to create an image of a fake person — though not a process that’s easy to replicate — raises concerns about misuse of the technology.

Editors' Recommendations

Hillary K. Grigonis
Hillary never planned on becoming a photographer—and then she was handed a camera at her first writing job and she's been…
How to hide photos on an Android phone or tablet
Google Photos

While today's best Android phones are quite secure as long as they remain locked with a passcode or biometrics like a fingerprint, by default those features only protect the front door. If someone picks up your phone while it's unlocked, there aren't typically any barriers that will keep them out of exploring everything from your contacts and emails to your photos.

This can be particularly challenging when it comes to photos, since those are the things we like to show off the most from our phones. We've likely all had those moments when we want to show a friend or co-worker a funny cat meme, so we hand over our phone and trust that they won't swipe right and see the photo of the hairy mole that we sent to our doctor that morning.

Read more
Best Sony A7 III deals: Save $300 on the full frame mirrorless camera
Sony A7 III

Sony A7 III Daven Mathies/Digital Trends / .

Even if you have one of the best camera phones at the moment, if you're a photography fanatic, you're probably feeling the burn for something new and exciting. For example, the Sony A7 III line is like none other in the world of handheld cameras and you can find several cameras in that line on sale right now at Best Buy. Here, we round up the best deals in the sale, then analyze how buying a Sony Alpha A7 III looks at the moment.
Today's best Sony A7 III deals

Read more
How to use (almost) any camera as a webcam for Zoom and more
how to use any camera as a web cam v2 00 58 01 still007

Cameras, even those in phones, brag about megapixels and lens specifications -- but laptops? Not so much. There’s a reason computer companies don’t say much about the webcams that come built into the bezels of their screens. Most of these cameras are low-quality, with tiny sensors and cheap lenses. Sure, they work for basic videoconferencing, but they aren’t very impressive and certainly leave us wanting something more.

While you could just buy a stand-alone webcam that connects over USB, to really take production value up a notch, you can opt for a DSLR or mirrorless camera. You’ll need a few workarounds to get this type of camera to be recognized as a webcam by your computer, but the trouble is worth it for the higher resolution, much better low-light performance, and cinematic background blur.

Read more