https://github.com/matlab-deep-learning/mtcnn-face-detection
Face detection and alignment using deep learning
https://github.com/matlab-deep-learning/mtcnn-face-detection
computer-vision deep-learning face-detection matlab mtcnn
Last synced: 4 months ago
JSON representation
Face detection and alignment using deep learning
- Host: GitHub
- URL: https://github.com/matlab-deep-learning/mtcnn-face-detection
- Owner: matlab-deep-learning
- License: other
- Created: 2020-01-14T11:12:04.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2021-07-01T20:04:09.000Z (almost 4 years ago)
- Last Synced: 2023-05-11T03:45:45.365Z (almost 2 years ago)
- Topics: computer-vision, deep-learning, face-detection, matlab, mtcnn
- Language: HTML
- Homepage:
- Size: 4.55 MB
- Stars: 25
- Watchers: 6
- Forks: 18
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Face Detection and Alignment MTCNN
[](https://codecov.io/gh/matlab-deep-learning/mtcnn-face-detection)
## [__Download the toolbox here__](https://github.com/matlab-deep-learning/mtcnn-face-detection/releases/latest/download/MTCNN-Face-Detection.mltbx)
This repository implements a deep-learning based face detection and facial landmark localization model using [multi-task cascaded convolutional neural networks (MTCNNs)](https://kpzhang93.github.io/MTCNN_face_detection_alignment/).
- [📦 Installation](#installation)
- [🏁 Getting Started](#getting-started)
- [🔎😄 Usage](#usage)
- [❓ About](#about)
- [💬 Contribute](#contribute)
_Note: This code supports inference using a pretrained model. Training from scratch is not supported. Weights are imported from the [original MTCNN model](https://kpzhang93.github.io/MTCNN_face_detection_alignment/) trained in Caffe._
## Installation
- Face Detection and Alignment MTCNN requires the following products:
- MATLAB R2019a or later (_now works in R2019a and later!_)
- Deep Learning Toolbox
- Computer Vision Toolbox
- Image Processing Toolbox
- Download the [latest release](https://github.com/matlab-deep-learning/mtcnn-face-detection/releases/) of the Face Detection and Aligment MTCNN. To install, open the .mltbx file in MATLAB.
## Getting Started
To get started using the pretrained face detector, import an image and use the `mtcnn.detectFaces` function:
```matlab
im = imread("visionteam.jpg");
[bboxes, scores, landmarks] = mtcnn.detectFaces(im);
```
This returns the bounding boxes, probabilities, and five-point facial landmarks for each face detected in the image.
## Usage
The `detectFaces` function supports various optional arguments. For more details, refer to the help documentation for this function by typing `help mtcnn.detectFaces` at the command window.
To get the best speed performance from the detector, first create a `mtcnn.Detector` object, then call its `detect` method on your image. Doing so ensures that the pretrained weights and options are loaded before calling detect:
```matlab
detector = mtcnn.Detector();
[bboxes, scores, landmarks] = detector.detect(im);
```
The detector object accepts the same optional arguments as the `mtcnn.detectFaces` function.
Refer to the MATLAB toolbox documentation or [click here](docs/getting_started.md) for a complete example.
## About
The MTCNN face detector is fast and accurate. Evaluation on the [WIDER face benchmark](http://mmlab.ie.cuhk.edu.hk/projects/WIDERFace/WiderFace_Results.html) shows significant performance gains over non-deep learning face detection methods. Prediction speed depends on the image, dimensions, pyramid scales, and hardware (i.e. CPU or GPU). On a typical CPU, for VGA resolution images, a frame rates ~10 fps should be achievable.
In comparisson to MATLAB's built in `vision.CascadeObjectDetector` the MTCNN detector is more robust to facial pose as demonstrated in the image below.
_Face detection from MTCNN in yellow, detections from the built in vision.CascadeObjectDetector in teal._
## Contribute
Please file any bug reports or feature requests as [GitHub issues](https://github.com/matlab-deep-learning/mtcnn-face-detection/issues). In particular if you'd be interested in training your own MTCNN network comment on the following issue: [Support training MTCNN](https://github.com/matlab-deep-learning/mtcnn-face-detection/issues/1)
_Copyright 2019 The MathWorks, Inc._