# lengstrom/fast-style-transfer

TensorFlow CNN for fast style transfer ⚡🖥🎨🖼

Repository: https://github.com/lengstrom/fast-style-transfer
Canonical: https://ross.abutalabs.com/products/fast-style-transfer
Language: Python
License Family: other
Topics: style-transfer, neural-style, neural-networks, deep-learning
Last push: 2023-07-16T02:36:30+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3695, "days_push": 1145, "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 10962, forks 2546 (observed 2026-08-28T04:10:44.865545+00:00)

## What it is
A TensorFlow implementation of fast neural style transfer that applies the style of famous paintings to photos and videos in real time. It is based on Johnson's perceptual losses and Ulyanov's instance normalization, and can stylize images in roughly 100ms on a Titan X GPU.

## Use cases
- apply painting styles to photos with a neural network
- stylize videos frame by frame
- run fast artistic style transfer on images
- train a custom style transfer model
- learn how perceptual loss style transfer works

## When to choose
- you want real-time or near-real-time style transfer rather than slow iterative optimization
- you need to stylize video, not just single images
- you want to train a network for a new painting style
- you are doing academic research on neural style transfer

## When to avoid
- you need a permissively licensed library for commercial use - the license restricts non-research use
- you want a maintained project on modern TensorFlow or PyTorch versions
- you need general image editing rather than artistic style transfer

## Facets
- artifact type: library
- maturity: maintenance
- function: image-processing, deep-learning, machine-learning
- domain: deep-learning, computer-vision, image-processing
- platform: python
- tags: style-transfer, tensorflow, neural-style, cnn, video-stylization, artistic-filters, gpu

## Member repositories
- lengstrom/fast-style-transfer (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:44.865545+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 2026-08-29T17:17:17.505096+00:00, confidence not recorded.
  - readme: https://github.com/lengstrom/fast-style-transfer (fetched 2026-08-28T04:10:44.865545+00:00, sha 36031a89fd57)
- Data as of 2026-08-30T08:39:29.467469+00:00.