search

Zhendong-Wang / Diffusion-GAN

Official PyTorch implementation for paper: Diffusion-GAN: Training GANs with Diffusion
MIT License
615 stars 65 forks source link

Generation includes reverse process of Diffusion? #39

Closed justin4ai closed 5 months ago

justin4ai commented 5 months ago

Hello, thanks for providing your amazing work.

I wonder if there is a reverse process in generate.py to generate images just like other diffusion models. Seemingly you only use GAN models for generation, so I just want to make whether my understand is proper clear!

Thanks!

Zhendong-Wang commented 5 months ago

Hi Justin,

Diffusion-GAN is motivated by using diffusion technologies to improve the training of GAN. It trains a one-step GAN generator.

Best,

Zhendong

On Sat, May 18, 2024 at 00:20 안준영 (Justin) @.***> wrote:

Hello, thanks for providing your amazing work.

I wonder if there is a reverse process in generate.py to generate images just like other diffusion models. Seemingly you only use GAN models for generation, so I just want to make whether my understand is proper clear!

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/Zhendong-Wang/Diffusion-GAN/issues/39, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKNPMGYT45X3GPQDPSFOM33ZC3JI5AVCNFSM6AAAAABH5D7FSGVHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDGOBUGA4DSMY . You are receiving this because you are subscribed to this thread.Message ID: @.***>

justin4ai commented 5 months ago

Dear Zhendong,

Thanks for making it clear! Then generation will be done via GAN, not via reverse process of diffusion model.

justin4ai commented 5 months ago

I close this issue