# genforce/interfacegan

[CVPR 2020] Interpreting the Latent Space of GANs for Semantic Face Editing

Repository: https://github.com/genforce/interfacegan
Canonical: https://ross.abutalabs.com/products/interfacegan
Homepage: https://genforce.github.io/interfacegan/
Language: Python
License: MIT
License Family: permissive
Last push: 2022-02-09T02:20:04+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2595, "days_push": 1667, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1561, forks 278 (observed 2026-08-28T04:05:03.890959+00:00)

## What it is
InterFaceGAN is a Python research library that interprets the latent space of pretrained GANs (PGGAN, StyleGAN) to find semantic subspaces for face editing. It provides pretrained attribute boundaries and scripts to manipulate facial attributes like age, gender, pose, and expression in generated faces.

## Use cases
- edit facial attributes in GAN-generated faces
- control age or gender in StyleGAN outputs
- interpret latent space of generative adversarial networks
- fix artifacts in GAN-synthesized images
- reproduce CVPR 2020 semantic face editing research
- change pose or expression in generated portraits

## When to choose
- you need latent-space semantic editing of faces with PGGAN or StyleGAN
- you are researching GAN interpretability and disentangled representations
- you want pretrained attribute boundaries for face manipulation

## When to avoid
- you need editing of non-face images or modern GANs beyond PGGAN/StyleGAN
- you need a production-ready image editing application
- you require active maintenance or recent framework compatibility

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, data-science
- domain: deep-learning, computer-vision, image-processing, artificial-intelligence
- platform: python, windows
- tags: gan, latent-space, face-editing, stylegan, pggan, pytorch, research, cvpr-2020, generative-models, linux, macos

## Member repositories
- genforce/interfacegan (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:03.890959+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:29:44.078250+00:00, confidence not recorded.
  - readme: https://github.com/genforce/interfacegan (fetched 2026-08-28T04:05:03.890959+00:00, sha bf283179a917)
  - homepage: https://genforce.github.io/interfacegan/ (fetched 2026-08-29T11:29:24.632224+00:00, sha c03e56e489d1)
- Data as of 2026-08-30T08:39:29.467469+00:00.