# NVlabs/FUNIT

Translate images to unseen domains in the test time with few example images.

Repository: https://github.com/NVlabs/FUNIT
Canonical: https://ross.abutalabs.com/products/funit
Homepage: https://nvlabs.github.io/FUNIT/
Language: Python
License: NOASSERTION
License Family: other
Last push: 2022-07-28T04:34:20+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": 2689, "days_push": 1497, "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 1586, forks 237 (observed 2026-08-28T04:05:07.721325+00:00)

## What it is
FUNIT is NVIDIA's PyTorch implementation of a few-shot unsupervised image-to-image translation model (ICCV 2019) that can translate images to unseen target domains specified by only a few example images at test time. It includes training and inference code plus an Animal Face dataset built from ImageNet crops.

## Use cases
- translate images to unseen domains with a few example images
- few-shot image-to-image translation with GANs
- reproduce FUNIT ICCV 2019 paper results
- train an image translation model on animal face datasets
- swap visual styles of animal species in photos

## When to choose
- you need few-shot image-to-image translation research code
- you want a reference PyTorch implementation of the FUNIT paper
- you have multi-GPU hardware for training GAN models

## When to avoid
- you need a commercially licensed model - code is CC BY-NC-SA 4.0, research only
- you lack high-end GPUs (paper results need 8x V100)
- you need actively maintained production software

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning, artificial-intelligence
- platform: python
- tags: gan, image-to-image-translation, few-shot-learning, pytorch, research-code, non-commercial-license, linux, gpu

## Member repositories
- NVlabs/FUNIT (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:07.721325+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 2026-08-30T03:55:27.302542+00:00, confidence not recorded.
  - readme: https://github.com/NVlabs/FUNIT (fetched 2026-08-28T04:05:07.721325+00:00, sha 1b9660b5bb7c)
  - homepage: https://nvlabs.github.io/FUNIT/ (fetched 2026-08-29T11:26:00.634928+00:00, sha 5ae8b425c564)
- Data as of 2026-08-30T08:39:29.467469+00:00.