The image-blending-using-gp-gans from msinghal34

msinghal34 / image-blending-using-gp-gans Goto Github PK

View Code? Open in Web Editor NEW

Inspired from GP-GAN: Towards Realistic High-Resolution Image Blending using PyTorch

Python 0.72% Jupyter Notebook 99.28%

image-blending-using-gp-gans's People

Stargazers

Watchers

image-blending-using-gp-gans's Issues

why the result is not smooth enough?

Pretrained model

Hey,

Is there any option to publish your tranied model?

Thanks!

Extension to new dataset

Hello, what is the format of files for re-training the network?

GP-GAN: error while blending with cpu.

Hi,
I ran the solution with this command

python3.6 blend.py -src test/src.png -dest dest.png -mask test/mask.png -model network_weights

and got this pytorch error:

raise RuntimeError('Attempting to deserialize object on a CUDA '
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

i know its due to cuda no available at my system but have installed cpu version of pytorch and also i saw that in your codes you took care of it with device = torch.device("cuda" if torch.cuda.is_available() else "cpu") which is good but still i'm facing this issue, so any help will be greatfull.
Thanks.