Paper Project

This is a unoffical implementation about Image Super-Resolution via Iterative Refinement by Pytorch.

Code will come soon.

• 16×16 -> 128×128 on FFHQ-CelebaHQ
• 64×64 -> 256×256 on FFHQ-CelebaHQ
• 64×64 -> 256×256 on ImageNet

• 1024×1024 face generation by a cascade of 3 models

1. Denoising Diffusion Probabilistic Models
2. Image Super-Resolution via Iterative Refinement