# VITA-Group/EnlightenGAN

[IEEE TIP] "EnlightenGAN: Deep Light Enhancement without Paired Supervision" by Yifan Jiang, Xinyu Gong, Ding Liu, Yu Cheng, Chen Fang, Xiaohui Shen, Jianchao Yang, Pan Zhou, Zhangyang Wang

Repository: https://github.com/VITA-Group/EnlightenGAN
Canonical: https://ross.abutalabs.com/products/enlightengan
Language: Python
License: NOASSERTION
License Family: other
Topics: unsupervised-learning, generative-adversarial-networks, gan, pytorch, low-light-enhance, low-light
Last push: 2024-04-25T07:46:42+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2635, "days_push": 860, "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 1091, forks 209 (observed 2026-08-28T04:03:33.055953+00:00)

## What it is
Official PyTorch implementation of EnlightenGAN, an IEEE TIP paper on deep low-light image enhancement without paired supervision using generative adversarial networks. It includes training and testing scripts, pretrained models, and unpaired training/testing datasets.

## Use cases
- enhance low-light photos without paired training data
- brighten dark images with a GAN model
- train an unsupervised light enhancement network
- run inference on dark images with pretrained EnlightenGAN weights
- reproduce the EnlightenGAN paper results
- compare low-light enhancement on LIME, MEF, NPE, VV, DICP benchmarks

## When to choose
- you need low-light image enhancement without paired supervision
- you want a research baseline or pretrained model for image relighting
- you have GPU resources and want to train or fine-tune a GAN-based enhancer

## When to avoid
- you need a production-ready, actively maintained image enhancement library
- you lack multi-GPU hardware and cannot adjust batch sizes
- you need paired-supervision training or real-time mobile inference

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, deep-learning, machine-learning
- domain: image-processing, computer-vision, deep-learning
- platform: python
- tags: gan, low-light-enhancement, unsupervised-learning, pytorch, research-code, ieee-tip, gpu, linux

## Member repositories
- VITA-Group/EnlightenGAN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:33.055953+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 2026-08-30T06:49:06.272059+00:00, confidence not recorded.
  - readme: https://github.com/VITA-Group/EnlightenGAN (fetched 2026-08-28T04:03:33.055953+00:00, sha f49309f22fc9)
- Data as of 2026-08-30T08:39:29.467469+00:00.