SpriteGAN

SpriteGAN generates Pokèmon sprites from original 151 Pokèmon Yellow sprites. It is a implementation of DCGAN using the keras library.

Deep Convolutional Generative Adversarial Network consists of two networks:

• Generator - tries to generate fake images from noise so as to fool the discriminator network
• Discriminator - tries to distinguish real images from fake ones

Dependencies

• Keras (Tensorflow Backend)
• Numpy
• ImageIO
• Jupyter

Usage

1. Download and install Jupyter Notebook and IPython kernel
2. run a Jupyter environment locally using jupyter notebook in the terminal
3. set path variable in model_train
4. set save location in save_imgs()
5. set use_prev to false to use pretrained models
6. call model_train for atleast 10 epochs to generate novel sprites

Example Output

References

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks