A Style-Based Generator Architecture for Generative Adversarial Networks [pdf]

Video: https://youtu.be/kSLJriaOumA?t=26

The quality and diversity of these images is incredible.

The "other datasets" section starting at 5:04 is very interesting. Aside from the disruptive effects of facial recognition, this tech seems destined to become a go-to tool for automotive design.

I believe that best of these generated faces will have a 90% match in training dataset.

If you go through the generated faces you can see all of them have different background. These are not generated from scratch, only picked from memory to match the requirements.

Each face has totally different hair and hairstyle.

Have the published the training dataset?

This is just phenomenal. Can see this being a a fairly disruptive force in the media industry.

Also, sock puppet factories could use this to create endless numbers of fake personas for social media astroturfing.

Watch this space..

The improvements in GANs from 2014 are amazing. From coarse 32x32 pixel images we have gotten to 1024x1024 images that can fool most humans.