A GAN implementation for generating low resolution fake faces.
In this project, I'll define and train a DCGAN on a dataset of faces.
I'll be using the CelebFaces Attributes Dataset (CelebA) to train the adversarial networks.
Since the project's main focus is on building the GANs, I'm using a dataset with some of the pre-processing done. Each of the CelebA images has been cropped to remove parts of the image that don't include a face, then resized down to 64x64x3 NumPy images. Some sample data is show below.
The dataset can be downloaded by clicking here
This is a zip file that needs to be extracted in the home directory of this notebook for further loading and processing. After extracting the data, you should be left with a directory of data processed_celeba_small/