# MobileFaceNet Face Landmark Detecter

*authors: David Wang*


## Desription

[MobileFaceNets](https://arxiv.org/pdf/1804.07573) is a class of extremely efficient CNN models to extract 68 landmarks from a facial image, which use less than 1 million parameters and are specifically tailored for high-accuracy real-time face verification on mobile and embedded devices. This repo is an adaptation from [cuijian/pytorch_face_landmark](https://github.com/cunjian/pytorch_face_landmark).

## Code Example

Extract facial landmarks from './img1.jpg'. 

*Write the pipeline in simplified style*:

```python
import towhee

towhee.glob('./img1.jpg') \
  .image_decode.cv2() \
  .face_landmark_detection.mobilefacenet() \
  .select('img','landmark') \
  .to_list()
```

*Write a same pipeline with explicit inputs/outputs name specifications:*

```python
import towhee

towhee.glob['path']('./img1.jpg') \
  .image_decode.cv2['path', 'img']() \
  .face_landmark_detection.mobilefacenet['img', 'landmark']() \
  .select('img','landmark') \
  .show()
```

<img src="https://towhee.io/face-landmark-detection/mobilefacenet/raw/branch/main/result.png" alt="result1" style="height:20px;"/>

## Factory Constructor

Create the operator via the following factory method

***face_landmark_detection.mobilefacenet(pretrained = True)***

**Parameters:**

​  ***pretrained***

​  whether load the pretrained weights.

​  supported types: `bool`, default is True, using pretrained weights.

## Interface

An image embedding operator takes an image as input. it extracts the embedding back to ndarray.

**Parameters:**

​	***img***: *towhee.types.Image (a sub-class of numpy.ndarray)*

​	The input image.


**Returns:** *numpy.ndarray*	

​	The extracted facial landmarks.