# AITTSMD/MTCNN-Tensorflow

Reproduce MTCNN using Tensorflow

Repository: https://github.com/AITTSMD/MTCNN-Tensorflow
Canonical: https://ross.abutalabs.com/products/mtcnn-tensorflow
Language: Python
License Family: other
Topics: mtcnn-tensorflow
Last push: 2019-12-16T14:30:14+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3289, "days_push": 2452, "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 1502, forks 703 (observed 2026-08-28T04:04:54.410381+00:00)

## What it is
A TensorFlow reproduction of MTCNN (Multi-task Cascaded Convolutional Networks) for joint face detection and facial landmark alignment. It includes the full training pipeline for the three-stage PNet, RNet, and ONet networks using WIDER Face and landmark datasets.

## Use cases
- detect faces in images with a cascaded CNN
- reproduce MTCNN training from the original paper
- train a face detection model on WIDER Face
- detect facial landmarks with TensorFlow
- generate hard examples for multi-stage face detector training
- run face detection and alignment in one pipeline

## When to choose
- you need to train or fine-tune MTCNN yourself rather than just run inference
- you want a faithful TensorFlow 1.x implementation of the MTCNN paper
- you need both face detection and landmark alignment from one model
- you are researching cascaded multi-task detection networks

## When to avoid
- you only need pretrained face detection inference with a modern maintained library
- your project uses TensorFlow 2.x or PyTorch
- you cannot provide CUDA GPUs for training
- you need production support or an active maintenance schedule

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, image-processing
- platform: python
- tags: mtcnn, face-detection, facial-landmarks, tensorflow, cascaded-networks, wider-face, celeba, linux, gpu

## Member repositories
- AITTSMD/MTCNN-Tensorflow (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:54.410381+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 2026-08-30T04:32:51.226734+00:00, confidence not recorded.
  - readme: https://github.com/AITTSMD/MTCNN-Tensorflow (fetched 2026-08-28T04:04:54.410381+00:00, sha fa8e5d8f68c2)
- Data as of 2026-08-30T08:39:29.467469+00:00.