Closed MetiCodePrivateLimited closed 3 years ago
Preserving finer details such as eyes and preserving details of complex backgrounds are some of the limitations of encoding into StyleGAN's latent representation. There is no quick solution to this issue but you could explore using other losses and encoders for better preservation of details. For keeping the background you could explore other post-processing techniques to extract only the face region and "paste" it to the original image. That way you get the original background untouched.
We trained model by using FFHQ images. Now, while applying SAM age effects on the image, it is changing the eye retina color. For more clarity, please see the image below:
Could you please guide me how to fix this issue i.e. to keep the color same after SAM age filters?
Also, the outer part of the face i.e. background is also changing. How can we fix the outer part of the face after applying SAM age filters?