GANnotation (PyTorch): Landmark-guided face to face synthesis using GANs (And a triple consistency loss!)
Switch branches/tags
Nothing to show
Clone or download
Latest commit 66cb6c5 Nov 13, 2018
Permalink
Failed to load latest commit information.
test_images Add files via upload Nov 12, 2018
GANnotation.py Add files via upload Nov 12, 2018
README.md Update README.md Nov 13, 2018
demo_gannotation.py Add files via upload Nov 12, 2018
model.py Add files via upload Nov 12, 2018
utils.py Add files via upload Nov 12, 2018

README.md

GANnotation

Face to face synthesis using Generative Adversarial Networks (model) (paper)

GANnotation example

This is the PyTorch repository for the GANnotation implementation. GANnotation is a landmark-guided face to face synthesis network that incorporates a triple consistency loss to bridge the gap between the input and target distributions

Release v1 (Nov. 2018): Demo showing the performance of our GANnotation

Release v2 will follow soon with the training code

Requirements

OpenCV --> pip install cv2 Link

PyTorch --> follow the steps in https://pytorch.org/

It also requires scipy and matplotlib, and the Python version to be 3.X

Use

The model can be downloaded from [https://drive.google.com/open?id=1YhpFXME3pnwy_WhgBtyhGhI1Y9WYYEOz]

Please, see the demo_gannotation.py file for usage

Contributions

All contributions are welcome

Citation

Should you use this code or the ideas from the paper, please cite the following paper:

@article{Sanchez2018Gannotation,
  title={Triple consistency loss for pairing distributions in GAN-based face synthesis},
  author={Enrique Sanchez and Michel Valstar},
  journal={arXiv preprint arXiv:1811.03492},
  year={2018}
}