# jcjohnson/fast-neural-style

Feedforward style transfer

Repository: https://github.com/jcjohnson/fast-neural-style
Canonical: https://ross.abutalabs.com/products/fast-neural-style
Language: Lua
License Family: other
Last push: 2023-10-03T20:54:31+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": 3620, "days_push": 1065, "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 4359, forks 801 (observed 2026-08-28T04:08:46.501056+00:00)

## What it is
A Torch (Lua) implementation of feedforward neural style transfer from the ECCV 2016 paper 'Perceptual Losses for Real-Time Style Transfer and Super-Resolution'. It includes pretrained models, training code, a real-time webcam demo, and an optimization-based Gatys-style transfer implementation.

## Use cases
- apply artistic style to images in real time
- stylize photos with pretrained neural style models
- train a custom feedforward style transfer model
- run a real-time webcam style transfer demo
- reproduce results from the perceptual losses paper
- implement instance normalization for stylization

## When to choose
- you need fast, feedforward image stylization with pretrained models
- you want to study or extend the ECCV 2016 perceptual losses method
- you are working in Torch/Lua and need style transfer code

## When to avoid
- you need a maintained PyTorch or TensorFlow implementation
- you require a permissively licensed project (no license is included)
- you want modern GPU stack support, since Torch is deprecated

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, deep-learning, machine-learning
- domain: computer-vision, image-processing, deep-learning, artificial-intelligence
- platform: lua, cpp
- tags: style-transfer, torch, neural-style, research-code, instance-normalization, computer-vision, gpu, linux, macos

## Member repositories
- jcjohnson/fast-neural-style (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:46.501056+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 2026-08-29T18:21:27.999684+00:00, confidence not recorded.
  - readme: https://github.com/jcjohnson/fast-neural-style (fetched 2026-08-28T04:08:46.501056+00:00, sha fc4df67202a5)
- Data as of 2026-08-30T08:39:29.467469+00:00.