Only watched the video, but one of the interesting things is the potential method to tell a generated image from a real one: namely, if you take a generated image, it's possible to find parameters which will generate exactly the same image. But if you take a real image, it's generally not possible to get exactly the same image, but only a similar one.
Comments
Only watched the video, but one of the interesting things is the potential method to tell a generated image from a real one: namely, if you take a generated image, it's possible to find parameters which will generate exactly the same image. But if you take a real image, it's generally not possible to get exactly the same image, but only a similar one.
The exact point in the video:
https://youtu.be/c-NJtV9Jvp0?t=208
Also, looks like https://thispersondoesnotexist.com/ has been updated to use the new generator.
Phew I looked at three and they all had toothy smiles where the teeth grew out out of the lips and one had a floating tooth.
This is only possible if you have access to the model neural net... If you dont you cant tell the difference.
Actually i was wrong. Generative Adverserial Nets often work accross Machine Learning models...