Implementation of Wasserstein Generative Adversarial Network (WGAN) with Gradient Penalty (GP) used for diseased fingerprint generation. The GAN is able to generate realisic looking images that look as though they are sampled from the same probability distribution as the training dataset. The repository includes the pre-trained models of generator and distriminator networks that were trained on a proprietary database of fingerprints afflicted by atopic eczema.
lilianedng / wgan-generation Goto Github PK
View Code? Open in Web Editor NEWWGAN-Generation: Generate synthetic diseased fingerprints with pre-trained GAN models, trained on an eczema fingerprint dataset.
License: MIT License