search

MedicineToken / MedSegDiff

Medical Image Segmentation with Diffusion Model
MIT License
1.06k stars 166 forks source link

The generated image is black #192

Open haoli-maker opened 1 month ago

haoli-maker commented 1 month ago

The image I generated after training looks like this, is that correct please? How can I process it to get the image in the author's paper? Thank you so much! image

MilkTeaAddicted commented 1 month ago

If you increase the contrast and brightness, you will find that there are generated contents inside. It seems that you might have used the wrong training model. Using the "emasavedmodel_0.9999_000000.pt" for training might be better.

MilkTeaAddicted commented 1 month ago

How many steps have you trained, and it may be caused by insufficient training time