1/ "ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks" is so packed with great insights. Of particular note: They pretrained their generator on L1 loss, and report that it actually improves quality. https://arxiv.org/abs/1809.00219
-
Show this thread
-
2/ I've been experimenting with pretraining both generator and critic with non-gan losses for DeOldify, it turns out, because I suspected it would lead to not only faster training but better results as well. It's still early but I'll just say it looks promising!
1 reply 0 retweets 4 likesShow this thread -
3/ The core concept definitely works functionally, I can tell you that much. Both for colorzation in DeOldify, as well as de-artifacting/super-res. This was part of lesson 7 in http://fast.ai V3 part 1, which will be released soon.
2 replies 1 retweet 3 likesShow this thread -
After pretraining, it's the same loss as before but then add critic loss in GAN training. That critic loss is what really makes it colorful, but it still needs to be constrained by the other losses (to replicate the picture, for one).
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.