-
Notifications
You must be signed in to change notification settings - Fork 29
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
FaceRestoreCF + GFPGAN 1.4 or 1.3 does not work after COMFYUI update #24
Comments
I got this error message: Failed inference for CodeFormer: ImageModelDescriptor.call() got an unexpected keyword argument 'w' |
use "codeformer.pth" then there will be no error |
@Big2State Thanks. It works. |
I know. I'm recovering old photos and have a huge number of comparisons between GFPGAN_v1.4 and codeformer. Codeformer changes facial features. He restores them incorrectly. The person in the photo no longer looks like himself. GFPGAN_v.1.4 also does not give 100% results, but in general the failure rate is somewhere around 10-15%. Recovery algorithms are much better than codeformer. Therefore, when someone gives advice to replace GFPGAN_v.1.4 with codeformer, this is most likely due to the fact that the neural network operator does not take into account the similarity of the photo, i.e. recognition of facial features from the original photo. |
@devZu9 Thanks for the feedback. I also noticed some changes on the faces, even color tones. But let's hope someone will fix this issue soon enough. |
I tried using codeformer.pth but still get this error "<urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1000)>" - all works fine in Automatic1111 |
I encountered the same issue |
does not work |
1.4不起作用 |
use: |
Hello.
I found a similar question in the ComfyUI discussion and they advised me to contact the developer FaceRestoreCF to resolve this issue.
I am not well versed in programming and therefore cannot tell you exactly what is causing the problem.
I can only show that now with GFPGAN we get white faces:
Please resolve the problem.
(I'm sorry, but I used an auto-translator into English, because I am not a native speaker of this language)
The text was updated successfully, but these errors were encountered: