Home>

What structure should I use when I want to learn that when I input an image with noise, an image with noise is output? Is it possible if U-Net is applied?
Also, I often see transfer learning in image recognition. Is transfer learning possible in image output?
Sorry for the basic question.