Creating Anime Faces using GANS
Given the https://www.kaggle.com/splcher/animefacedataset dataset. I have created a DCGAN network to generate images of the faces of anime characters.
GAN Architecture invloves the use of two neural networks. A discriminator network which tells apart AI generated images from real images and a generator network which given the feedback from the discriminator network learns to generate a more and more realistic image.
Diagram of GANs architecture:
Real Images:
AI Generated Images after 1 epoch:
AI generated Image after 25 epochs:
Model Losses:
