#176 · Primary category: Image Generation
Cartoon-StyleGAN
Fine-tuning StyleGAN2 for Cartoon Face Generation
Project last updated:10/16/21
GitHub Stars
655
Forks
120
Contributors
2
License
Other
Why we included this project
Turning a real face photo into a consistent cartoon or webtoon character is a common ask, and this repo shows a workable path using a pretrained StyleGAN2. The author fine-tunes StyleGAN2-ADA on small, unbalanced datasets such as webtoon faces, and introduces two concrete techniques: freezing the low-resolution layers of the generator and the style vector (FreezeSG), plus a structure loss, to keep the source face's identity and pose while shifting the style. Pretrained models for Disney, webtoon, and Metfaces are included, and the notebook demonstrates downstream edits like changing expression or pose, even text-driven changes via StyleCLIP-style latent optimization. The Colab notebook lets you cartoonize your own face without setting up a full training environment, which makes it handy for artists, hobbyists, and teams prototyping stylized avatar pipelines.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
stable-diffusion-webui
Stable Diffusion web UI
ComfyUI
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
diffusers
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.
upscayl
🆙 Upscayl - #1 Free and Open Source AI Image Upscaler for Linux, MacOS and Windows.
InvokeAI
InvokeAI is a leading open-source creative engine for Stable Diffusion, offering an industry-leading web UI for generating and editing visual media.