🧠 Facial Recognition System – Built by Logical Hammad

Created with ❤️ by Logical Hammad — this AI-powered web app uses face recognition, emotion detection, and user registration to deliver an accurate, secure, and intelligent experience.
It’s a full-stack system built using Python, Flask, face_recognition, and DeepFace — with a sleek, responsive frontend to make everything feel simple and smooth.

---

🧠 Introduction

This is a smart AI-powered Facial Recognition System built using Python, Flask, and cutting-edge libraries like face_recognition and DeepFace.
Created with love by Logical Hammad, this project aims to bring powerful recognition and classification capabilities into a sleek, intuitive web interface.

---

🚀 Technologies Used

• Frontend: HTML, CSS, JavaScript
• Backend: Flask (Python)
• AI Libraries:
  • face_recognition – for face encoding and matching
  • deepface – for emotion, gender, and age detection
• Database: File-based using .npy (NumPy arrays)

---

🧑‍💼 Registration System

The system has a user-friendly registration flow that ensures high-quality data collection for reliable face recognition.

📸 Image Upload Requirements

Each user must upload six images from different angles:

1. Front - Straight
2. Front - Straight (Smiling)
3. Left - 30°
4. Left - 45°
5. Tilt Up
6. Tilt Down

These ensure robust encoding by capturing facial features from multiple viewpoints.

✅ Drag-and-drop or click-to-upload supported
✅ Preview of uploaded images with placeholders
✅ Real-time error display under each card for:

• Image resolution and file type
• Face visibility and detection
• Multiple faces or no face
• Blurriness or bad lighting

Once passed, face encodings are generated via face_recognition and stored using NumPy .npy format, along with the user's metadata.

---

🕵️ Recognition Flow

To recognize a person, simply upload an image.
The system performs:

• Frontend & backend validation
• Face encoding generation
• Matching using Euler distance
• If match found below the threshold → ✅ Identity confirmed
• If no perfect match → 🤔 Shows top possible matches

Additionally, using deepface, the system guesses:

• 🧠 Mood
• 🎂 Age
• 🚻 Gender

---

🎨 User Interface

This web app features a clean, modern, and responsive UI built to enhance usability while looking professional.
Everything from drag-and-drop uploads, animated cards, to dynamic error messages is thoughtfully designed for the end user.

---

📂 Folder Structure

facial-recognition/
├── database
    └── username
        └── Their encodings in .npy file
├── static
    ├── CSS
    ├── JS
    └── Images
├── templates
    └── common
└── uploads

---

💡 Future Improvements

• Admin panel to manage users
• Live camera-based detection
• Database shift to MongoDB or Firebase
• API-based integration into other platforms
• Upload via mobile camera

---

🧪 Try It Yourself

Clone the repo, set up a virtual environment, and install dependencies:

git clone https://github.com/logicalhammad/Facial-Recognition

---

🎥 Related YouTube Video

📺 [Coming Soon] Watch how this system works — a full walkthrough on the Logical Hammad YouTube channel

🔗 About the Creator I'm Hammad Mustafa, also known as Logical Hammad — a software engineer, AI enthusiast, and co-founder of Nelston Technologies. I build smart tech with purpose — from IoT to AI and everything in between.

This was Logical Hammad. I think you’ll love this project. 🔗 An Awesome Project