search

suxuann / ddib

Dual Diffusion Implicit Bridges for Image-to-Image Translation. ICLR 2023.
MIT License
357 stars 30 forks source link

Instructions on creating my own dataset? #14

Open chokevin8 opened 1 year ago

chokevin8 commented 1 year ago

Are there any instructions on creating my own dataset and starting to train from scratch (rather than loading pretrained model). Also do you think the image to image translation (Style transfer) would work well in medical images? I'm trying to use histopathological images (H&E) images. Also, like others have said, when would the color translation model be released? I'm really looking forward to applying this.

I would appreciate a response. Thank you!

xiayhh commented 1 year ago

May I ask if you have used this model to train your own dataset of medical images, I am also working on this.

chokevin8 commented 1 year ago

@xiayhh Unfortunately, I haven't tried yet, still hoping that the authors of the repository @suxuann @chenlin9 could release instructions and also the color translation model...

LVCHENYONG commented 1 year ago

Hello, I am also doing image to image conversion, I would like to ask if you found the method of ddib, I found some problems in the SyntheticModel framework