myRen

> There is no mention of discriminator in the paper. Where did you find it? Hi，is there a discriminator which works in you method/code?

> In code, there is a discriminator. You can see how to enable it on vox-adv.yaml I noticed that you do not use discriminator in default. How do you think...

> It is not that useful in my experience. THANKS！

> It is not that useful in my experience. I want the first-order-motion model to run on the iphone, so i change the parameters in self-256-generator.xml to reduce the inference...

> No. The author replied in #97. Only the used 5 attributes are supported. You have to retrain the model yourself for other attributes. Is there some idea about how...

> You can use the faces with closes eyes to train this model linking is https://github.com/zhangqianhui/GazeCorrection Thanks！The DeepWarp is your baseline model, could you share the implementation of DeepWarp. Very...

What is the result if we do not use .cuda() ?

> I download coco_resnet_50_map_0_335.pt from https://drive.google.com/open?id=1hCtM35R_t6T8RJVSd74K4gB-A1MR-TxC ,when load using torch.load(" coco_resnet_50_map_0_335.pt") it raise error : > result = unpickler.load() > ModuleNotFoundError: No module named 'model' > > what is the...

> @yunjey @youngjung > First, I want to thank you for your good works including the high-quality paper and very organized codes. > StarGANv2 can generate realistic synthetic images that...

> There is only the 100000_nets_ema.ckpt, but I want to continue to train based on the pre-trained model, including 100000_net.ckpt and 100000_optims.ckpt (actually mapping network, style encoder and discriminator) Is...