# timesler/facenet-pytorch

Pretrained Pytorch face detection (MTCNN) and facial recognition (InceptionResnet) models

Repository: https://github.com/timesler/facenet-pytorch
Canonical: https://ross.abutalabs.com/products/facenet-pytorch
Language: Python
License: MIT
License Family: permissive
Topics: pytorch, resnet, inception-resnet, face-recognition, face-identification, face-tracking, mtcnn, pretrained-models, face-detection, facial-recognition
Last push: 2025-09-16T22:04:42+00:00

## Health v2 (maintenance only)
Score: 42/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 42, release rhythm 8, longevity 100
- inputs: {"age_days": 2658, "days_push": 351, "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 5162, forks 1000 (observed 2026-08-28T04:09:11.506140+00:00)

## What it is
A PyTorch library providing pretrained face detection (MTCNN) and facial recognition (Inception ResNet V1) models, ported from the TensorFlow facenet project. It includes one of the fastest MTCNN implementations available and supports complete detection-to-embedding pipelines.

## Use cases
- detect and crop faces from images
- generate face embeddings for identity matching
- build a face login or verification system
- track faces across video frames
- cluster photos by person
- finetune face recognition models on new data

## When to choose
- you need pretrained face detection and recognition in PyTorch
- you want a fast, pure-PyTorch MTCNN implementation
- you need face embeddings compatible with VGGFace2/CASIA-Webface models
- you want a simple pip-installable pipeline from image to embedding

## When to avoid
- you need state-of-the-art accuracy on hard faces (consider newer detectors like RetinaFace or ArcFace models)
- you work outside PyTorch (e.g., TensorFlow or ONNX-only pipelines)
- you need non-face object detection or general image recognition
- you require active large-scale development - the project is mature and changes infrequently

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, computer-vision, image-processing, deep-learning
- domain: computer-vision, machine-learning, image-processing, artificial-intelligence
- platform: python, cross-platform
- tags: face-recognition, face-detection, mtcnn, inception-resnet, pretrained-models, pytorch, facial-embeddings, vggface2, gpu

## Member repositories
- timesler/facenet-pytorch (main) score 42

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:11.506140+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 2026-08-29T18:01:53.619050+00:00, confidence not recorded.
  - readme: https://github.com/timesler/facenet-pytorch (fetched 2026-08-28T04:09:11.506140+00:00, sha 4dc0e9cfa30d)
  - registry_pypi: https://pypi.org/pypi/facenet-pytorch/json (fetched 2026-08-29T08:55:55.490256+00:00, sha 9dd4ae75b8f0)
- Data as of 2026-08-30T08:39:29.467469+00:00.