swapping-autoencoder-pytorch

swapping-autoencoder-pytorch copied to clipboard

Hi, great work! But I have some questions about 1024 resolution. In the paper:  could you provide the detail about the implemention of smaller network capacity?  And In...

hughwcq

some detail about the code

2

Haosouth

pretrained model

1

hello, the link to pretrained models can not be opened

PengQiTu

structure not being retained

5

hi @taesungp great piece of work, I trained it on my dataset of 50k images for 50 Mil iterations as you suggested, on testing time the results are quite impressive...

haiderasad

how to keep the structure and extract feature?

4

Your work is so amazing. When I use the interpolation in the texture code, it will change the structure of source picture slightly. In the mountain dataset, for example, we...

hughwcq

About Finetune

3

Hi,great work! I just finetuned the officially mountain checkpoints you provided,but there were many artifacts.So may I ask you about your train setting or some training details?

hughwcq

tom99763

Region editing

4

Hi @taesungp , great project ! I have a question that in your paper and demo video you show that the model can change the region in the object using...

kudoNCT

Does the algorithm work with providing texture images for swapping?

1

Hello! Congratulations for your very impressive work! I was wondering if the same results may be achieved in the case when we provide images of different kinds of textures that...

SeptimiaSarbu

some details about the paper and project

1

Hi,it`s a amazing work！But there are some details i want to know. In the paper:"A manipulation vector for snow is discovered by taking mean difference between 10 user-collected photos of...

hughwcq