search

zsyzzsoft / co-mod-gan

[ICLR 2021, Spotlight] Large Scale Image Completion via Co-Modulated Generative Adversarial Networks
Other
444 stars 67 forks source link

Data Prep #26

Open tiwarikaran opened 3 years ago

tiwarikaran commented 3 years ago

Can you explain how the dataset was prepared for Co Mod GAN's like the pix2pix requires the photos to be indexed left being the input right being the output, like that, how is the dataset prepared here. And what should I do to apply this on a custom dataset

zsyzzsoft commented 3 years ago

If you want to run image completion experiments, you only need to put all the images in a folder and run our provided script. If you want to run image-to-image translation experiments, the data preparation code is not ready now but maybe I'll release the code in the next week.

tiwarikaran commented 3 years ago

okay, no no I am just interested in Inpainting, but facing issues in colaab

hongruihuang commented 10 months ago

Hey,is there any instruction for data preparation to run image-to-image translation on the custom dataset?