# ipazc/mtcnn

MTCNN face detection implementation for TensorFlow, as a PIP package.

Repository: https://github.com/ipazc/mtcnn
Canonical: https://ross.abutalabs.com/products/mtcnn
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: mtcnn, face, detection, tensorflow, pip, package, python3, landmark
Last push: 2024-10-08T16:09:34+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": 3162, "days_push": 694, "days_rel": 695, "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 2485, forks 527 (observed 2026-08-28T04:06:55.503179+00:00)

## What it is
A Python library implementing the MTCNN (Multitask Cascaded Convolutional Networks) algorithm for face detection and facial landmark alignment, built on TensorFlow 2.x and distributed as a pip package. It detects faces with bounding boxes and predicts keypoints (eyes, nose, mouth) using a cascade of three neural networks, with support for batch processing and device selection.

## Use cases
- detect faces in images with bounding boxes
- find facial landmarks like eyes, nose and mouth
- align faces before face recognition preprocessing
- batch process a folder of photos to extract face crops
- filter images that contain no detectable faces
- build a face verification pipeline in Python

## When to choose
- you need accurate face detection plus landmark points in a Python/TensorFlow stack
- you want a simple pip-installable detector with a minimal API
- you need face alignment keypoints for downstream recognition or embedding models
- you want batch processing of images on CPU or GPU

## When to avoid
- you need real-time video face detection at high frame rates on edge devices
- your stack is PyTorch-only and you don't want a TensorFlow dependency
- you need face recognition or identity matching, not just detection
- you need masks, occlusion handling, or state-of-the-art detector accuracy on hard cases

## Facets
- artifact type: library
- maturity: stable
- function: computer-vision, image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, machine-learning
- platform: python, cross-platform
- tags: face-detection, facial-landmarks, mtcnn, tensorflow, face-alignment, bounding-boxes, pip-package, gpu

## Member repositories
- ipazc/mtcnn (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:55.503179+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 2026-08-30T02:28:04.125056+00:00, confidence not recorded.
  - readme: https://github.com/ipazc/mtcnn (fetched 2026-08-28T04:06:55.503179+00:00, sha 40907b492097)
  - registry_pypi: https://pypi.org/pypi/mtcnn/json (fetched 2026-08-29T10:10:09.116294+00:00, sha 75d691a8477f)
- Data as of 2026-08-30T08:39:29.467469+00:00.