BriefGPT.xyz
Dec, 2021
CLIP2StyleGAN:无监督提取 StyleGAN 编辑方向
CLIP2StyleGAN: Unsupervised Extraction of StyleGAN Edit Directions
HTML
PDF
Rameen Abdal, Peihao Zhu, John Femiani, Niloy J. Mitra, Peter Wonka
TL;DR
本文提出一种方法将StyleGAN和CLIP的预训练潜空间有效链接,从而能够自动从StyleGAN中提取语义标注的编辑方向,找到并命名有意义的编辑操作,而不需要任何额外的人类指导
Abstract
The success of
stylegan
has enabled unprecedented
semantic editing
capabilities, on both synthesized and real images. However, such editing operations are either trained with semantic supervision or described usi
→