PhotoFaceDetect

Table of Contents

• Introduction
• Features
• Architecture
• Installation
  • Prerequisites
  • Clone the Repository
  • Install Dependencies
  • Download Pre-trained Models
• Usage
  • Run Script
  • Available Scripts
• Workspace Structure
• Models
• License
• Contributing

Introduction

PhotoFaceDetect is a Python-based application designed for real-time face detection, tracking, and analysis. Leveraging powerful libraries like OpenCV and YOLOv3-tiny, this project offers multiple scripts to cater to various face detection and processing needs. Whether you're capturing faces from your desktop screen or streaming from a webcam, PhotoFaceDetect provides efficient and customizable solutions.

Features

• Real-Time Face Detection: Utilize OpenCV's DNN module and YOLOv3-tiny for efficient face detection.
• Multithreading Optimization: Enhance performance using Python's concurrent.futures.
• Customizable Scripts: Multiple scripts available for different face detection approaches.
• Model Integration: Supports various pre-trained models for age and gender detection.
• Automated Photo Capture: Save detected faces as separate image files with ease.

Installation

Prerequisites

• Python 3.8 or higher
• pip package manager

Clone the Repository

git clone https://github.com/mbn-code/PhotoFaceDetect.git
cd PhotoFaceDetect

Install Dependencies

Ensure all required Python packages are installed

pip install -r requirements.txt

Download Pre-trained Models

The project relies on several pre-trained models located in the models directory. Ensure you have the following files:

• deploy.prototxt
• res10_300x300_ssd_iter_140000.caffemodel
• yolov3-tiny.cfg
• yolov3-tiny.weights
• coco.names

You can download these models from their respective official sources:

• YOLOv3-tiny
• Caffe Models
• COCO Names

Place the downloaded files in the models directory.

Usage

Run Script

To start the application, use the provided run.sh script. This script allows you to select and execute different face detection and processing scripts.

./run.sh

Available Scripts

The run.sh script provides the following options:

• Run faceMeshPro_desk_GPU

  • Executes: src/faceMeshPro_desk_GPU.py
  • Description: Captures frames from the desktop screen and performs face detection using GPU acceleration.

• Run faceMesh

  • Executes: faceMesh.py
  • Description: Utilizes YOLOv3-tiny for real-time face detection from a webcam stream.

• Run faceDeteID

  • Executes: faceDeteID.py
  • Description: Detects faces and performs ID tracking with real-time display.

• Run faceDeteIDphoto

  • Executes: faceDeteIDphoto.py
  • Description: Detects faces from a video stream and saves them as photos.

• Run faceDetePro

  • Executes: faceDetePro.py
  • Description: Advanced face detection with multiple processing layers.

• Run faceDetePro2

  • Executes: faceDetePro2.py
  • Description: Enhanced face detection with skin color analysis.

• Run faceMesh_pro_video

  • Executes: faceMesh_pro_video.py
  • Description: Processes video streams for professional-grade face detection.

• Run YOLOv3-tiny Face Detection

  • Executes: src/yolov3_tiny_face_detection.py
  • Description: Implements YOLOv3-tiny for efficient face detection.

• Exit

  • Exits the application.

Upon running, you'll be prompted to select an option:

========================================
       Welcome to the Face Detection App       
========================================

Please select an option:
1) Run faceMeshPro_desk_GPU
2) Run faceMesh
3) Run faceDeteID
4) Run faceDeteIDphoto
5) Run faceDetePro
6) Run faceDetePro2
7) Run faceMesh_pro_video
8) Run YOLOv3-tiny Face Detection
0) Exit
Enter your choice:

Workspace Structure

Project Directory Structure

PhotoFaceDetect/
├── assets/
│ └── images/
├── coco/
│ └── coco.names
├── models/
│ ├── age_net.caffemodel
│ ├── deploy_age.prototxt
│ ├── deploy_gender.prototxt
│ ├── deploy.prototxt
│ ├── gender_net.caffemodel
│ ├── model_CNN_V2.h5
│ ├── res10_300x300_ssd_iter_140000.caffemodel
│ ├── yolov3-tiny.cfg
│ └── yolov3-tiny.weights
├── src/
│ ├── faceDeteID.py
│ ├── faceDeteIDphoto.py
│ ├── faceDetePro.py
│ ├── faceDetePro2.py
│ ├── faceMesh_pro_desk_GPU.py
│ ├── faceMesh_pro_video.py
│ ├── faceMesh.py
│ └── ...
├── .gitignore
├── README.md
├── requirements.txt
└── run.sh

Directory Descriptions

• assets/: Contains static assets like images used in the project.
• coco/: Holds the COCO class labels file (coco.names).
• detected_faces/: Directory where detected face images are stored.
• face_photo/: Stores saved face photos.
• models/: Contains pre-trained models and their configuration files.
• src/: Source code for various face detection and processing scripts.
• .gitignore: Specifies files and directories to be ignored by Git.
• README.md: Project documentation.
• requirements.txt: Lists Python dependencies.
• run.sh: Bash script to execute different scripts based on user input.

Models

The models directory includes several pre-trained models essential for face detection and analysis:

Caffe Models

• deploy.prototxt
• res10_300x300_ssd_iter_140000.caffemodel

YOLOv3-tiny

• yolov3-tiny.cfg
• yolov3-tiny.weights

Age and Gender Detection

• deploy_age.prototxt
• deploy_gender.prototxt
• age_net.caffemodel
• gender_net.caffemodel
• model_CNN_V2.h5

Ensure all models are properly placed in the models directory for the scripts to function correctly.

License

This project is licensed under the MIT License.

Contributing

Contributions are welcome! Please follow these steps:

1. Fork the Repository
2. Create a Feature Branch

git checkout -b feature/YourFeature

1. Commit Your Changes

git commit -m "Your Message"

git push origin feature/YourFeature

1. Open PR