Facial Recognition: Detector 2D (Python API)

• Introduction
• Running program
• Requirements
• Status
• Related projects

Introduction

facialRecognitionDetection2D module use dlib python API. The module detects faces using pre-trained models and adds facial recognition doing a facial training with user images. Also use YARP to send video source pre and post-procesed. Also admits YARP source video like input. Some scripts added to create automated files of user image database and user name database. This module also publish detection results in YARP port.

Documentation available on docs

Running program

facialRecognitionDetection2D requires images of people to detect. Images should be located in database dir. The process to create database files:

1. Execute scripts/configureDatabase.py, to create image index.

python3 configureDatabase.py

2. Execute programs/facialRecognitionDetection2D.py, to start de program.

python facialRecognitionDetection2D.py

3. Connect video source to facialRecognitionDetection2D.

yarp connect /videoSource /facialRecognitionDetection2D/img:i

NOTE:

• Video results are published on /facialRecognitionDetection2D/img:o
• Data results are published on /facialRecognitionDetection2D/data:o

Requirements

facialRecognitionDetection2D requires:

• Install OpenCV 3.0.0+
• Install YARP 2.3.XX+
• Install pip
• Install face_recognition:

(Using OpenCV and YARP with Python 2.7 bindings)

pip2 install face_recognition

(Using OpenCV and YARP with Python 3 bindings)

pip3 install face_recognition

Tested on: ubuntu 14.04, ubuntu 16.04, ubuntu 18.04, lubuntu 18.04 , raspbian, Kubuntu 20.04 and Windows 10.

Status

Related projects

• ageitgey: face_recognition project
• davidvelascogarcia: Tensorflow: Lite Detector 2D (Python API)