Recognize faces Application

Table of contents:

• Introduction
• Project Features
• Project Structure
• Quick Preview
• Requirements to run
• Run the project
• Try a demo
• Contributors

---

Introduction

Our application, built with C++, specializes in face detection and recognition using PCA/Eigen analysis. We aim to provide reliable and efficient methods for detecting and recognizing faces in both color and grayscale images. By utilizing C++, we ensure top-notch performance and accurate results.

---

Project Features

In this project:

• Live Face Detection:
  • Our application is equipped with real-time face detection functionality. It can instantly detect faces as they appear in camera feeds, providing instantaneous results for dynamic environments.
• Face Detection:
  • Our application employs advanced techniques to detect faces in images. It can analyze both color and grayscale images, making it versatile and adaptable to various scenarios. The face detection algorithm efficiently identifies facial features, enabling accurate localization of faces within the images.
• Face Recognition with PCA/Eigen Analysis:
  • Once the faces are detected, our application utilizes PCA/Eigen analysis for face recognition. This method extracts the essential facial features and represents them in a lower-dimensional space. By comparing these features, the application can recognize and match faces with a high degree of accuracy.
• Performance Reporting:
  • Our application provides comprehensive performance reports, allowing users to evaluate the effectiveness of the face detection and recognition algorithms. The reports include metrics such as precision, providing insights into the application's performance.
• ROC Curve Plotting:
  • In addition to performance metrics, our application generates Receiver Operating Characteristic (ROC) curves. These curves plot the true positive rate against the false positive rate at various thresholds, providing a visual representation of the algorithm's performance. ROC curves are widely used in evaluating the effectiveness of face recognition systems.

---

Project Structure

The Application is built using:

• Qt Version 6.4.2
• OpenCV 4.9.0

main
├─ FaceRecognition
│  ├─ FaceRecognition/Models:
│  ├─ FaceRecognition/Views: include ui/header / resource file related to pages design.
│  ├─ FaceRecognition/Controllers: Contains the controllers responsible for handling communication between models and views.
│  ├─ FaceRecognition/Helpers: used for common app logic and different ui functionality.
│  ├─ FaceRecognition/Assets:  Models used in facedetection
│  ├─ FaceRecognition/data: generated .txt files saving the trained PCA
│  ├─ FaceRecognition/dataset: dataset used in training/testing
└─ Gallery/test/unknowns : images used for testing.

---

Quick Preview

Face Detection

Offline Face Recgonition

Face Recgonition 1	Face Recgonition 2	Face Recgonition 3

Face Recgonition 4	Face Recgonition 5

Live Face Detection and Recognition

---

Requirements to run

Qt Setup and openCV

1. Ensure Qt is installed on your system.
2. Clone the VisionX repository to your local machine.

---

Installation

How to run the project

• Download and Install Qt from the online installer using minGW 64 architecture.

www.qt.io

• Download and build openCV 4.9.0 from their website

www.opencv.org/releases/

• if you don't know how to build OpenCV on your PC please refer to the installCV.md

• Go to the Task1.pro file and ensure that openCV libraries are referred to in your pc like that.

Windows

INCLUDEPATH += <yourCV build folder path>\install\include

LIBS += -L<yourCV build folder path>\release\lib \
    -lopencv_calib3d490                 \
    -lopencv_core490                    \
    -lopencv_features2d490              \
    -lopencv_flann490                   \
    -lopencv_highgui490                 \
    -lopencv_imgproc490                 \
    -lopencv_imgcodecs490               \
    -lopencv_photo490                   \
    -lopencv_stitching490               \
    -lopencv_ts490                      \
    -lopencv_video490                   \
    -lopencv_videoio490                 \
    -lopencv_objdetect490               \

Linux

CONFIG += link_pkgconfig
PKGCONFIG += opencv4

Building

1. Open the project in Qt Creator.
2. Make sure your .pro file matches commonpro.txt file
3. Add folders "dataset4" with the training dataset, "dataset4files" empty , "test" with your testing data, "dataset4live" with training live data to your build directory
4. Build the project to generate the executable.

Try a demo

[ Download Here !]

---

Contributors

Abdulrahman Emad

Mourad Magdy

Ziad Meligy

Mariam Ahmed

Youssef Ashraf