ermongroup / SDEdit

PyTorch implementation for SDEdit: Image Synthesis and Editing with Stochastic Differential Equations
https://sde-image-editing.github.io/
MIT License
966 stars 90 forks source link

CelebA-HQ pth missing #9

Open ucalyptus2 opened 3 years ago

ucalyptus2 commented 3 years ago

I could not find for CelebA-HQ in https://github.com/ermongroup/SDEdit/blob/main/functions/process_data.py @junyanz

aferriss commented 3 years ago

+1 I'd also love to try out the celeba model.

chenlin9 commented 3 years ago

Thank you for the question.

You can directly load the CelebA-HQ model by using "—config celeba.yml". However, you will need to create your own mask and image, upload the file to the "colab_demo directory", and then change "—npy_name" argument to be the name of the file. More details can be found at https://github.com/ermongroup/SDEdit/blob/main/runners/image_editing.py#L108 We are currently working on an interactive interface for creating masks and editing images.

Let me know if you have any other questions. Thank you.

sorayutmild commented 3 years ago

@chenlin9 How can we create our own mask and image. Is it .pth file?

sorayutmild commented 3 years ago

I've already solved it. just using torch.save() for example. torch.save([test_mask, test_img_tensor], './colab_demo/test_image.pth')

xinzi2018 commented 2 years ago

How can you create CelebA-HQ mask?

inbarhub commented 1 year ago

Hi, Can you supply the images from celebA you edit in the paper?

hxy-123-coder commented 1 year ago

@xinzi2018 Hello,do you solve this problem?