Noise2noise assumes access to the uncorrupted images during the learning phase. Why not use a supervised method in this case? This one however does not: https://openreview.net/forum?id=BJg4Z3RqF7 …
Nvidia’s “Noise 2 Noise” paper is a great read because it really stretches your thinking about what neural nets are actually doing and the possibilities. Also it happens to be potentially relevant to DeOldify! Paper here: https://arxiv.org/abs/1803.04189
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.