https://github.com/woctezuma/steam-stylegan2-ada

Train a StyleGAN2-ADA model on Colaboratory to generate Steam banners.
https://github.com/woctezuma/steam-stylegan2-ada

colab colab-notebook colaboratory gan generative-adversarial-network google-colab google-colab-notebook google-colaboratory steam steam-api steam-data steam-game steam-games steam-gan steam-pics steam-store stylegan stylegan-model stylegan2 stylegan2-ada

Last synced: 19 days ago
JSON representation

Train a StyleGAN2-ADA model on Colaboratory to generate Steam banners.

• Host: GitHub
• URL: https://github.com/woctezuma/steam-stylegan2-ada
• Owner: woctezuma
• License: mit
• Created: 2020-10-10T12:00:13.000Z (about 4 years ago)
• Default Branch: main
• Last Pushed: 2021-02-01T12:03:14.000Z (almost 4 years ago)
• Last Synced: 2023-03-11T16:58:06.342Z (almost 2 years ago)
• Topics: colab, colab-notebook, colaboratory, gan, generative-adversarial-network, google-colab, google-colab-notebook, google-colaboratory, steam, steam-api, steam-data, steam-game, steam-games, steam-gan, steam-pics, steam-store, stylegan, stylegan-model, stylegan2, stylegan2-ada
• Language: Jupyter Notebook
• Homepage: https://colab.research.google.com/
• Size: 14.4 MB
• Stars: 31
• Watchers: 3
• Forks: 8
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Steam StyleGAN2-ADA

The goal of this [Colab][colab-website] 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][data-repository],

-   Run [`StyleGAN2_ADA_training.ipynb`][colab-notebook-training] to train a StyleGAN2-ADA model from scratch.

[![Open In Colab][colab-badge]][colab-notebook-training]

-   Run [`StyleGAN2_ADA_image_sampling.ipynb`][colab-notebook-sampling] to generate images with a trained StyleGAN2-ADA model,

[![Open In Colab][colab-badge]][colab-notebook-sampling]

-   To automatically resume training from the latest checkpoint, you will have to use [my fork][stylegan2-ada-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`][download_steam_banners].

[![Open In Colab][colab-badge]][download_steam_banners]

Images were then filtered (duplicates, outliers, etc.) with [`remove_duplicates.ipynb`][filter_steam_banners].

[![Open In Colab][colab-badge]][filter_steam_banners]

## References

-   DCGAN:

    -   [Radford, Alec, et al. *Unsupervised Representation learning with Deep Convolutional GAN*. ICLR 2016.][dcgan-paper]

    -   [Official implementation][dcgan-official-repository]

    -   [Application to Steam banners][dcgan-applied-to-steam-banners]

-   StyleGAN:

    -   [Karras, Tero, et al. *A Style-Based Generator Architecture for Generative Adversarial Networks*. CVPR 2019.][stylegan1-paper]

    -   [Official implementation][stylegan1-official-repository]

    -   [Application to Steam banners][stylegan1-applied-to-steam-banners]

-   StyleGAN2:

    - [Karras, Tero, et al. *Analyzing and Improving the Image Quality of StyleGAN*. CVPR 2020.][stylegan2-paper]

    -   [Official implementation][stylegan2-official-repository]

    -   [Application to Steam banners][stylegan2-applied-to-steam-banners]

-   StyleGAN2-ADA:

    -   [Karras, Tero, et al. *Training generative adversarial networks with limited data*. NeurIPS 2020.][stylegan2-ada-paper]

    -   Official implementations: [TensorFlow][stylegan2-ada-official-repository] and [PyTorch][stylegan2-ada-pytorch-repository]

    -   [Application to Steam banners][stylegan2-ada-applied-to-steam-banners]

[download_steam_banners]: 

[filter_steam_banners]: 

[colab-website]: 

[colab-notebook-training]: 

[colab-notebook-sampling]: 

[colab-badge]: 

[data-repository]: 

[stylegan2-ada-fork]: 

[dcgan-paper]: 

[stylegan1-paper]: 

[stylegan2-paper]: 

[stylegan2-ada-paper]: 

[dcgan-official-repository]: 

[stylegan1-official-repository]: 

[stylegan2-official-repository]: 

[stylegan2-ada-official-repository]: 

[stylegan2-ada-pytorch-repository]: 

[dcgan-applied-to-steam-banners]: 

[stylegan1-applied-to-steam-banners]: 

[stylegan2-applied-to-steam-banners]: 

[stylegan2-ada-applied-to-steam-banners]: