Are the training parameters provided by you the same as those on the website (https://cleanup.pictures/) ? about lama

[windj007]

Hi!

Honestly, I do not know which model is used in Cleanup Pictures. What images and outputs do you compare?

Why do the same pictures and masks have different effects?

This might be also due to different input resolution - Cleanup Pictures downscale input image to 720p, if I'm not mistaken. Feeding same input image and mask but varying input resolution will definitely result in different outputs.

[zhouwy19]

What resolution this repo code will resize the input image to?

[windj007]

By default, our predict.py does not resize image at all - the image is fed into the generator as is