# cchen156/Learning-to-See-in-the-Dark

Learning to See in the Dark. CVPR 2018

Repository: https://github.com/cchen156/Learning-to-See-in-the-Dark
Canonical: https://ross.abutalabs.com/products/learning-to-see-in-the-dark
Homepage: http://cchen156.web.engr.illinois.edu/SID.html
Language: Python
License: MIT
License Family: permissive
Topics: python, rawpy, tensorflow
Last push: 2025-09-15T19:03:26+00:00

## Health v2 (maintenance only)
Score: 51/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 42, release rhythm 35, longevity 100
- inputs: {"age_days": 3072, "days_push": 352, "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 5565, forks 837 (observed 2026-08-28T04:09:22.413348+00:00)

## What it is
TensorFlow implementation of 'Learning to See in the Dark' (CVPR 2018), a deep learning model that brightens very dark, short-exposure RAW photos. It includes training/testing code, pretrained models, and the See-in-the-Dark Sony/Fuji datasets.

## Use cases
- enhance dark low-light photos with deep learning
- train a low-light RAW image enhancement model
- denoise and brighten short-exposure RAW images
- reproduce CVPR 2018 See-in-the-Dark results
- download the SID Sony and Fuji benchmark datasets
- run inference on near-black RAW photos

## When to choose
- you need research-grade low-light RAW image enhancement
- you want to benchmark against the SID dataset
- you're working with TensorFlow and RAW files via rawpy

## When to avoid
- you need a maintained production pipeline (Python 2.7 / TF 1.x era code)
- you want a simple photo app without GPU setup
- you can't handle large 25-52 GB dataset downloads

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, computer-vision
- domain: computer-vision, image-processing, deep-learning, machine-learning
- platform: python
- tags: low-light-image-enhancement, raw-image-processing, tensorflow, cvpr-2018, computational-photography, research-code, see-in-the-dark-dataset, linux, gpu

## Member repositories
- cchen156/Learning-to-See-in-the-Dark (main) score 51

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:22.413348+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 2026-08-29T17:55:37.073923+00:00, confidence not recorded.
  - readme: https://github.com/cchen156/Learning-to-See-in-the-Dark (fetched 2026-08-28T04:09:22.413348+00:00, sha 7c5ed02f4a39)
  - homepage: http://cchen156.web.engr.illinois.edu/SID.html (fetched 2026-08-29T08:50:48.408835+00:00, sha a84fcc50b15a)
- Data as of 2026-08-30T08:39:29.467469+00:00.