/Image2StyleGAN

This is an implementation of the paper Image2StyleGAN using StyleGAN as backbone

Primary LanguagePython

Image2StyleGAN

This is an implementation of the paper Image2StyleGAN using StyleGAN-v1 as the backbone

Acknowledgement

This project is a part of my internship at King Adbullah University of Science and Technology(KAUST) under the supervision of Professor Peter Wonka

References

The code for the architecture was taken from https://github.com/pacifinapacific/StyleGAN_LatentEditor and the official TensorFlow weights were converted to Pytorch weights to be used in the project. This code uses the Official StyleGAN architecture from NVIDIA

Pre-trained Model

Here are the pre-trained weights in TensorFlow and PyTorch (Converted from TensorFlow weights): Google Drive

This project makes use of PyTorch weights. If you need any additional information, please contact me.

Experimental Results

Screenshot Screenshot Screenshot Screenshot Screenshot

Stay tuned for some more results on various experimental results.