search

glucauze / sd-webui-faceswaplab

Extended faceswap extension for StableDiffusion web-ui with multiple faceswaps, inpainting, checkpoints, ....
https://glucauze.github.io/sd-webui-faceswaplab/
GNU Affero General Public License v3.0
731 stars 97 forks source link

version 1.2.0 is not as good as 1.1.2 ，the swaped face is very blur, and more complex operation #25

Closed glucauze closed 1 year ago

glucauze commented 1 year ago

version 1.2.0 is not as good as 1.1.2 ，the swaped face is very blur, and more complex operation

Originally posted by @sdcarterchen in https://github.com/glucauze/sd-webui-faceswaplab/discussions/23#discussioncomment-6640076

glucauze commented 1 year ago

@sdcarterchen

version 1.2.0 is not as good as 1.1.2 ，

No. There is no reason, the processing is the same. It can be simpler if configured.

and more complex operation

Yes. If you don't care about the options in faces, don't use them. It will be back to very simple.

the swaped face is very blur :

Yes. I was fearing that comment.

I disabled codeformer by default in Global Postprocessing. That's why you get poor results. Activate it and that will be back to what you know. You can make it the default in global settings.

Here are back to 1.1.2 default settings. Codeformer will be enabled for all faces (including non swapped -- default 1.1.2 behavior)

default_settings

You can ignore det_size since it's not in this version.

Here are my advanced default settings I only use codeformer for faces (and LDSR or 003_realSR_BSRGAN_DFOWMFC_s64w8_SwinIR-L_x4_GAN only on faces)

my_settings

NOTE ! : Restart SD after changing these settings.

sdcarterchen commented 1 year ago

Thank you very much. These settings are very useful to me.