Smart Attendance System (Face Recognition)

A production-oriented face recognition–based attendance system with:

• Admin-controlled sessions
• Real-time attendance marking
• No student login (face = identity)
• Clean separation of backend, frontend, and camera client

This repository is designed so it can be cloned, deployed, and extended for real institutions.

---

System Overview

Actors

• Admin: Controls sessions, views attendance, analytics
• Student: No login, identified via face recognition

Flow

1. Admin logs into dashboard
2. Admin starts an attendance session
3. Camera client recognizes faces
4. Backend validates active session
5. Attendance is marked once per session

---

Tech Stack

Backend

• FastAPI
• SQLAlchemy
• PostgreSQL
• Session-based admin authentication

Frontend

• Next.js (App Router)
• Role-protected admin pages
• API-driven UI

Face Recognition Client

• OpenCV
• Haar Cascade
• Face encodings (local)
• Camera-based recognition

---

📁 Project Structure

├── Backend/ # API & business logic
│ ├── routes/ # Auth, attendance, sessions, analytics
│ ├── dependencies/ # Role guards & auth helpers
│ ├── models.py # DB models
│ ├── schemas.py # Request/response schemas
│ ├── session_manager.py # Attendance session control
│ └── main.py # FastAPI entry point
│
├── Client/ # Camera & face recognition
│ ├── capture_faces.py
│ ├── encode_faces.py
│ └── recognize_attendance.py
│
├── frontend/ # Admin dashboard (Next.js)
│
├── dataset_/ # Student face images
├── encodings_/ # Generated face encodings
└── README.md

---

⚙️ Prerequisites

• Python 3.9+
• Node.js 18+
• PostgreSQL
• Webcam (USB or built-in)

---

Backend Setup

cd Backend
python -m venv venv
source venv/bin/activate   # Windows: venv\Scripts\activate
pip install -r requirements.txt

Environment Variables

# Create .env using .env.example:

DATABASE_URL=postgresql://user:password@localhost/attendance_db
SECRET_KEY=your_secret_key

# Run Backend
uvicorn main:app --reload

API will be available at:

http://localhost:8000

Admin Setup

Seed initial admin user. And use the seeded credentials to log in.

python scripts/seed_admin.py

Frontend Setup

cd frontend
npm install
npm run dev

Open:

http://localhost:3000

Admin-only routes are protected via backend auth.

Face Recognition Setup

Step 1: Capture Student Faces

cd Client
python capture_faces.py

1. Enter student ID
2. Capture multiple angles
3. Images saved in dataset_/

Step 2: Encode Faces

python encode_faces.py

1. Converts images into numerical encodings
2. Stored in encodings_/

Step 3: Start Attendance Recognition

python recognize_attendance.py

What this does:

• Opens camera
• Detects faces
• Matches encodings
• Sends attendance to backend
• Prevents duplicate marking

Attendance Session Logic

Attendance is only marked when:

• Admin session is ACTIVE

• Student is recognized

• Student has not already been marked

This logic lives in:

Backend/session_manager.py

How to Extend This Project

You can:

• Add attendance reports (PDF / CSV)
• Add multi-class or multi-branch support
• Replace Haar Cascade with deep learning models
• Add hardware camera integration
• Deploy backend + frontend on cloud, keep client local

The architecture is intentionally modular.

Notes

Face recognition runs locally (not as an API)

Backend only validates and stores data

This avoids security and performance issues

AND IF YOU STILL GET STUCK AND NEED HELP SETTING UP THIS PROJECT REACH OUT TO ME @sanc0o0 on linkedin or drop me an email at sanansari0305@gmail.com