harskish/ganspace

Discovering Interpretable GAN Controls [NeurIPS 2020]

48
/ 100
Emerging

Applies PCA in activation space to discover interpretable edit directions across GAN layers, enabling fine-grained semantic controls like viewpoint and lighting without retraining. Supports BigGAN, StyleGAN, and StyleGAN2 with both Z and W latent spaces, featuring interactive exploration tools and layer-wise edit composition for precise image synthesis manipulation.

1,797 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

1,797

Forks

266

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Jan 20, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/harskish/ganspace"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.