Physicist91/dcgan

Simple Implementation of Deepfake with Keras and TensorFlow

Deep Convolutional Generative Adversarial Network

Main resources

• Papers: DCGAN, GAN
• TensorFlow: [DCGAN]https://www.tensorflow.org/tutorials/generative/dcgan, CNN

In a GAN:

• The Generator takes a vector of random numbers and outputs an image.
• The Discriminator classifies the image given as input into either "real" or "fake".

In a CNN:

• The number of conv2d filters correspond to the dimension of the output space.