This video explores changes to the StyleGAN architecture to remove certain artifacts, increase training speed, and achieve a much smoother latent space interpolation! This paper also presents an interesting Deepfake detection algorithm enabled by their improvements to latent space interpolation.
Robert Luxemburg’s StyleGAN2 Interpolation Loop:
Paper Link:
Tero Karras StyleGAN Video:
NVIDIA StyleGAN Repository:
Thanks for watching! Please subscribe!