# Shawn-Shan/fawkes

Fawkes, privacy preserving tool against facial recognition systems. More info at https://sandlab.cs.uchicago.edu/fawkes

Repository: https://github.com/Shawn-Shan/fawkes
Canonical: https://ross.abutalabs.com/products/fawkes
Homepage: https://sandlab.cs.uchicago.edu/fawkes
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: privacy-protection, privacy-enhancing-technologies, face-recognition, adversarial-machine-learning
Last push: 2023-08-02T16:26:04+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2299, "days_push": 1127, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5595, forks 510 (observed 2026-08-28T04:09:22.756759+00:00)

## What it is
Fawkes is a privacy protection tool from University of Chicago researchers that adds imperceptible adversarial perturbations to photos to prevent unauthorized facial recognition models from identifying you. It is distributed as a Python CLI on PyPI with low/mid/high protection modes.

## Use cases
- cloak my photos against facial recognition
- protect personal images from face recognition models
- add adversarial perturbations to photos before posting online
- prevent unauthorized face identification from scraped images
- research adversarial machine learning defenses against face recognition

## When to choose
- you want to protect personal photos from being used to train or run facial recognition models
- you need a simple CLI to batch-process image directories with cloaking perturbations
- you are studying adversarial ML and privacy-enhancing technologies

## When to avoid
- you need GPU acceleration out of the box - the packaged binary is CPU-only and GPU requires modifying the source
- you need real-time or fast processing - generation takes about 60 seconds per image on CPU
- you need a guarantee of protection - effectiveness against evolving models is not assured

## Facets
- artifact type: cli-tool
- maturity: maintenance
- function: image-processing, machine-learning, privacy, cli
- domain: privacy, machine-learning, image-processing, security
- platform: python, cli, cross-platform
- tags: facial-recognition, adversarial-examples, cloaking, face-recognition, privacy-protection, gpu

## Member repositories
- Shawn-Shan/fawkes (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:22.756759+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 2026-08-29T17:55:29.296850+00:00, confidence not recorded.
  - readme: https://github.com/Shawn-Shan/fawkes (fetched 2026-08-28T04:09:22.756759+00:00, sha 0aca9e939bcf)
  - registry_pypi: https://pypi.org/pypi/fawkes/json (fetched 2026-08-29T08:50:22.807370+00:00, sha b29ce3111279)
- Data as of 2026-08-30T08:39:29.467469+00:00.