clip2latent: Text driven sampling of a pre-trained StyleGAN using denoising diffusion and CLIP.

Justin N. M. Pinkney, Chuan Li

Browse the full BMVC paper archive.