Have you try more deeper networks?

[zhangqianhui]

Hello , Soumith As we all know , vgg , inception , resnet , this networks are very deep And Have you tried deeper networks to contruct the gans? For example Using reset in D networks.

[soumith]

We've used existing VGG networks in our workshop paper "Semantic Segmentation using Adversarial Networks", because it was a fine-tuning task rather than generation from scratch.

[zhangqianhui]

@soumith thanks for your answer , but when to open the code ?

[soumith]

i dont know yet.

[zhangqianhui]

@soumith

https://github.com/imatge-upc/saliency-salgan-2017

use vgg in Generative networks