retinaface/README.md

RetinaFace Face Detection

Authors: David Wang

Desription

This operator detects faces in the images by using RetinaFace Detector[1]. It will return the bounding box positions and the confidence scores of detected faces. This repo is a adapataion from [2].

Code Example

Load an image from path './dog.jpg' and use the pretrained RetinaFace to generate face bounding boxes.

Write the pipeline in simplified style:

from towhee import ops

bboxes = dc.glob('nmb46.jpg') \
  .image_decode.cv2() \
  .face_detection.retinaface() \
  .to_list() 

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

from towhee import dc

dc.glob['path']('./dog.jpg') \
  .image_decode.cv2['path', 'img']() \
  .image_embedding.timm['img', 'vec'](model_name='resnet50') \
  .face_detection.retinaface() \
  .to_list()

Factory Constructor

Create the operator via the following factory method.

ops.face_detection.retinaface()

Interface

A face detection operator takes an image as input. it generates the bounding box positions and confidence scores back to ndarray.

Parameters:

​ image: numpy.ndarray.

​ the image to detect faces.

​ supported types: numpy.ndarray

Returns:

numpy.ndarray

​ The detected face bounding boxes.

numpy.ndarray

​ The detected face bounding boxes confident scores.

Reference

[1]. https://arxiv.org/abs/1905.00641
[2]. https://github.com/biubug6/Pytorch_Retinaface