
Now you will not surprise anyone with filters that improve the quality of the photos. But the restoration of old portraits still leaves something to be desired. Older photos tend to be too blurred, so normal image hook methods will not work on it.
The SomantCarc team has developed a new algorithm to improve the old photo portraits based on a combination of two generative adversary network methods (GFP-GAG).
The first method is trying to isolate any digital noise and improve facial areas on the photo, while the second method tries to reject the unsuccessful options generated by the first.
According to the researchers, this approach makes it possible to obtain a quality of restoration of a blind face compared to previous methods and algorithms.
Anyone can assess the work of scientists and try to restore old snaps to https://huggingface.co/spaces/akhaliq/gfpgan
More details on the GFP-Gan method can be found to https://github.com/tencentarc/gfpgan
