# 1adrianb/face-alignment

:fire: 2D and 3D Face alignment library build using pytorch

Repository: https://github.com/1adrianb/face-alignment
Canonical: https://ross.abutalabs.com/products/face-alignment
Homepage: https://www.adrianbulat.com
Language: Python
License: BSD-3-Clause
License Family: permissive
Topics: python, deep-learning, face-alignment, face-detector, pytorch, face-detection
Last push: 2026-04-06T22:19:17+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 76, release rhythm 46, longevity 100
- inputs: {"age_days": 3274, "days_push": 149, "days_rel": 149, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7538, forks 1383 (observed 2026-08-28T04:10:00.625281+00:00)

## What it is
A Python library built on PyTorch that detects 2D and 3D facial landmarks in images using the FAN deep learning face alignment network. It supports multiple pluggable face detectors (SFD, BlazeFace, YuNet, RetinaFace, SCRFD) and batch processing of directories.

## Use cases
- detect facial landmarks in photos
- extract 3D face keypoints from images
- face alignment preprocessing for face recognition pipelines
- batch process a folder of images for face landmarks
- swap face detector backends for speed vs accuracy
- run face landmark detection on GPU or CPU

## When to choose
- you need accurate 2D or 3D facial landmark detection in Python
- you want a PyTorch-based, pip-installable face alignment solution
- you need to choose between accuracy- and speed-optimized face detectors

## When to avoid
- you need real-time video landmark tracking at high frame rates on CPU
- you need full face recognition or identification rather than landmark detection
- you work outside the Python/PyTorch ecosystem

## Facets
- artifact type: library
- maturity: active
- function: computer-vision, image-processing, machine-learning, deep-learning
- domain: computer-vision, deep-learning, image-processing
- platform: python, cross-platform
- tags: facial-landmarks, face-detection, pytorch, 2d-3d-landmarks, face-alignment, gpu

## Member repositories
- 1adrianb/face-alignment (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:00.625281+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 2026-08-29T17:37:26.180146+00:00, confidence not recorded.
  - readme: https://github.com/1adrianb/face-alignment (fetched 2026-08-28T04:10:00.625281+00:00, sha 7eec7534292c)
  - homepage: https://www.adrianbulat.com (fetched 2026-08-29T08:32:34.153353+00:00, sha 75ec4a72efa6)
  - site_page: https://www.adrianbulat.com/about-me (fetched 2026-08-29T08:32:34.191349+00:00, sha 0448aa938c2f)
  - registry_pypi: https://pypi.org/pypi/face-alignment/json (fetched 2026-08-29T08:32:34.194017+00:00, sha 457043adf0d6)
- Data as of 2026-08-30T08:39:29.467469+00:00.