Steam StyleGAN2-ADA

The goal of this Colab notebook is to capture the distribution of Steam banners with a StyleGAN2-ADA model.

Usage

• Acquire the data, e.g. as a snapshot called 256x256.zip in my data repository,
• Run StyleGAN2_ADA_training.ipynb to train a StyleGAN2-ADA model from scratch.
• Run StyleGAN2_ADA_image_sampling.ipynb to generate images with a trained StyleGAN2-ADA model,
• To automatically resume training from the latest checkpoint, you will have to use my fork of StyleGAN2-ADA.

Data

The dataset consists of 14k Steam banners with RGB channels and resized from 300x450 to 256x256 resolution.

Images were downloaded with download_steam_banners.ipynb.

Images were then filtered (duplicates, outliers, etc.) with remove_duplicates.ipynb.

References

• DCGAN:
  • Radford, Alec, et al. Unsupervised Representation learning with Deep Convolutional GAN. ICLR 2016.
  • Official implementation
  • Application to Steam banners
• StyleGAN:
  • Karras, Tero, et al. A Style-Based Generator Architecture for Generative Adversarial Networks. CVPR 2019.
  • Official implementation
  • Application to Steam banners
• StyleGAN2:
  • Karras, Tero, et al. Analyzing and Improving the Image Quality of StyleGAN. CVPR 2020.
  • Official implementation
  • Application to Steam banners
• StyleGAN2-ADA:
  • Karras, Tero, et al. Training generative adversarial networks with limited data. NeurIPS 2020.
  • Official implementations: TensorFlow and PyTorch
  • Application to Steam banners