# EvgenyKashin/stylegan2-distillation

Repository: https://github.com/EvgenyKashin/stylegan2-distillation
Canonical: https://ross.abutalabs.com/products/stylegan2-distillation
Homepage: https://arxiv.org/abs/2003.03581
Language: Python
License: NOASSERTION
License Family: other
Last push: 2022-02-02T18:55:59+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": 2371, "days_push": 1673, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1018, forks 102 (observed 2026-08-28T04:03:14.823856+00:00)

## What it is
A research implementation of the ECCV 2020 paper 'StyleGAN2 Distillation for Feed-forward Image Manipulation', distilling StyleGAN2 latent-space edits into fast paired image-to-image networks (pix2pixHD). It provides pretrained weights and notebooks for face transformations like gender swap, aging, and style mixing.

## Use cases
- swap gender in face photos
- make a face look older or younger
- apply style transfer to portraits
- morph between face images
- run fast feed-forward image editing instead of slow latent optimization
- generate synthetic paired training data with StyleGAN2

## When to choose
- you need fast feed-forward face manipulation comparable to StyleGAN2 latent optimization quality
- you want pretrained pix2pixHD weights for gender swap or aging
- you are reproducing the ECCV 2020 paper or researching GAN distillation

## When to avoid
- you need a maintained production tool or commercial license (nonstandard license, little recent activity)
- you want general-purpose image editing beyond the specific face transformations provided
- you lack a GPU or the stylegan2/pix2pixHD dependency stack

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, image-processing, deep-learning
- domain: computer-vision, image-processing, deep-learning, artificial-intelligence
- platform: python
- tags: gan, stylegan2, pix2pixhd, face-manipulation, image-to-image-translation, distillation, eccv-2020, research-code, gpu

## Member repositories
- EvgenyKashin/stylegan2-distillation (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:14.823856+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 2026-08-30T07:09:25.781738+00:00, confidence not recorded.
  - readme: https://github.com/EvgenyKashin/stylegan2-distillation (fetched 2026-08-28T04:03:14.823856+00:00, sha 0d0aa6a423f9)
  - homepage: https://arxiv.org/abs/2003.03581 (fetched 2026-08-29T13:09:49.812853+00:00, sha 0729e3b1f8f2)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T13:09:49.815726+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T13:09:49.820812+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T13:09:49.823174+00:00, sha a1f16f915a9a)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T13:09:49.818196+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.