AI-Based Face Recognition Attendance System

Progressive Learning - 7 Minggu Pembelajaran Terstruktur

Sistem absensi berbasis face recognition yang dikembangkan secara progressive dalam 7 minggu pembelajaran terstruktur. Menggunakan MediaPipe Face Detection + Face Mesh untuk performa real-time yang cepat dan akurat.

---

🎯 Fitur Utama

• ✅ MediaPipe Face Recognition: Detection + Face Mesh (1404-dim landmarks)
• ✅ Super Fast: 60+ FPS real-time (0.01s per face)
• ✅ High Accuracy: Cosine similarity matching
• ✅ Lightweight: ~150MB install (no TensorFlow!)
• ✅ File-Based Storage: Pickle + JSON + CSV (no database needed!)
• ✅ Real-time attendance tracking via webcam
• ✅ Desktop GUI dengan Tkinter
• ✅ Export reports ke CSV/JSON
• ✅ Support 100-200+ persons
• ✅ Berjalan offline - no internet needed

---

🚀 Quick Start

Prerequisites

1. Python 3.11+

python --version  # Harus 3.11 atau lebih

Download: https://python.org/downloads/ (Centang "Add to PATH"!)

2. Webcam - Built-in atau USB camera

Installation

# 1. Clone/Download project
git clone <repo-url>
cd ExtraQueensya

# 2. Create virtual environment
python -m venv venv
venv\Scripts\activate  # Windows
# source venv/bin/activate  # Mac/Linux

# 3. Install dependencies
pip install -r requirements.txt

# 4. Verify
python -c "import cv2, mediapipe; print('✅ Ready!')"

Setup per-week dengan Interactive Menu:

# Week 4-7: Auto-setup dataset
cd minggu-X-xxxxx
python setup_weekX.py
# Interactive menu akan muncul dengan pilihan:
#   [1] Copy dataset dari week sebelumnya (otomatis!)
#   [2] Capture faces dengan camera
#   [3] Skip (populate manual nanti)
# Tinggal ketik nomor, no manual commands needed!

Setup All Weeks (4-7) Sekaligus:

python setup_all_weeks.py
# Create folder structure untuk semua week
# Kemudian run setup_weekX.py untuk interactive dataset population

---

📚 Learning Path (7 Minggu)

minggu-1-python-basics/       ← START: OpenCV basics
├── learning/                 ← Tutorial & teori
├── project/                  ← Implementation & tests
└── README.md

minggu-2-face-detection/      ← Face detection (MediaPipe)
minggu-3-face-recognition/    ← Face recognition (MediaPipe Face Mesh)
minggu-4-dataset-database/    ← Dataset management
minggu-5-recognition-system/  ← Recognition service integration
minggu-6-attendance-system/   ← Attendance logic & reports
minggu-7-desktop-gui/         ← Desktop application

Cara Belajar

Setiap minggu:

1. Baca learning/README.md → Pahami konsep
2. Jalankan tutorial files di learning/ → Praktik
3. Baca project/README.md → Understand implementation
4. Run tests di project/ → Validasi

Timeline: 2-3 jam/hari × 7 minggu = Complete attendance system

---

🏗️ Tech Stack

Core:

• Python 3.11 - Programming language
• OpenCV 4.8.1 - Image processing & video capture
• MediaPipe 0.10.9 - Face Detection + Face Mesh
  • Face Detection: min_detection_confidence=0.3 (configurable)
  • Face Mesh: 468 landmarks × 3 coordinates = 1404-dim encoding
• NumPy 1.26.2 - Cosine similarity matching

Storage & Export:

• Pickle (face encodings - 1404-dim vectors)
• JSON (metadata)
• CSV (attendance logs)
• Tkinter (GUI)

Why MediaPipe?

• ✅ Super fast: 60+ FPS real-time (25x faster than DeepFace)
• ✅ Lightweight: ~150MB install vs ~500MB with TensorFlow
• ✅ No heavy dependencies
• ✅ Easy setup: 2-3 min install
• ✅ Google-maintained, production-ready

---

📖 Week-by-Week Roadmap

Week	Topic	Files	Output	Time
1	Python & OpenCV	5 tutorials	image_utils.py	6-7 hari
2	Face Detection	4 tutorials	face_detector.py	6-7 hari
3	Face Recognition	3 tutorials	face_recognizer.py (MediaPipe)	7-8 hari
4	Dataset Management	3 tutorials	dataset_manager.py	6-7 hari
5	System Integration	2 tutorials	recognition_service.py	6-7 hari
6	Attendance System	2 lessons	attendance_system.py	6-7 hari
7	Desktop GUI	2 lessons	main_app.py	5-6 hari

Total: ~50 hari pembelajaran terstruktur

---

🎮 Running the Application

Week 1-3: Learning Mode

cd minggu-X-xxxxx/learning
python lesson-X/main.py

Week 4+: Project Mode

# Setup dataset first
cd minggu-X-xxxxx
python setup_weekX.py

# Run learning
cd learning/lesson-1
python main.py

Week 7: Desktop App

cd minggu-7-desktop-gui/project
python main_app.py

---

❓ FAQ

Q: Harus mulai dari Week 1?
A: Pemula → Ya. Intermediate → Skip ke Week 3. Advanced → Week 5.

Q: Berapa lama per minggu?
A: Flexible! 2-3 jam/hari santai, atau 1-2 hari intensif.

Q: Kenapa pakai MediaPipe?
A: Super fast (60+ FPS), lightweight (~150MB), no heavy dependencies, Google-maintained.

Q: Bisa offline?
A: Ya! Semua berjalan lokal, no cloud/internet needed.

Q: Face detection tidak akurat?
A: Check lighting, distance (50-100cm), frontal face. Atau tuning confidence di Week 6 README.

Q: Cara tuning sensitivity/confidence?
A: Lihat section "⚙️ Configuration & Tuning" di minggu-6-attendance-system/README.md.

---

📂 Progressive Modules Architecture

Setiap minggu standalone dengan copy modules dari minggu sebelumnya:

• Week 1: image_utils.py
• Week 2: Week 1 + face_detector.py
• Week 3: Week 2 + face_recognizer.py (MediaPipe Face Mesh)
• Week 4: Week 3 + dataset_manager.py
• Week 5: Week 4 + recognition_service.py
• Week 6: Week 5 + attendance_system.py
• Week 7: Week 6 + GUI (main_app.py)

Why Duplicate? Student bisa langsung run tanpa import issues atau dependency ke folder lain. Setiap week bisa dibuka sebagai standalone project.

---

🎯 Expected Results

Setelah 7 minggu, kamu akan punya:

• ✅ Working attendance system dengan face recognition
• ✅ Desktop application dengan GUI
• ✅ Super fast real-time recognition (60+ FPS)
• ✅ Report export (CSV/JSON)
• ✅ Portfolio-ready project
• ✅ Understanding of confidence tuning & optimization
• ✅ Skills: Python, OpenCV, MediaPipe, AI, Computer Vision, GUI

---

📞 Resources

Documentation

• Main README (ini) - Overview & quick start
• Weekly READMEs - 14 detailed READMEs (learning + project per minggu)
• KUNCI_JAWABAN_TUGAS.md - Answer keys untuk tugas (after trying!)
• Setup Scripts - setup_week4.py sampai setup_week7.py dengan interactive menus
• Configuration Guide - Week 6 README section "⚙️ Configuration & Tuning"

External

• OpenCV Docs: https://docs.opencv.org/
• MediaPipe Docs: https://google.github.io/mediapipe/
• MediaPipe Face Detection: https://google.github.io/mediapipe/solutions/face_detection
• MediaPipe Face Mesh: https://google.github.io/mediapipe/solutions/face_mesh

Need Help?

1. Baca README di week yang sedang dikerjakan
2. Run python setup_weekX.py untuk interactive setup
3. Check Week 6 README untuk confidence tuning
4. Google error message
5. Review tutorial files di learning/

---

💪 Tips untuk Sukses

1. Consistency > Intensity - 2 jam/hari lebih baik dari 14 jam/weekend
2. Understand > Memorize - Pahami konsep, jangan hafal code
3. Test Everything - Run test setelah setiap perubahan
4. Take Notes - Catat konsep penting & insights
5. Build Portfolio - Screenshot, demo video, write blog
6. Ask for Help - Stuck > 1 jam? Take break atau cari bantuan

---

🚀 Ready to Start?

Complete Beginner:

cd minggu-1-python-basics/learning
python lesson-1/main.py

Intermediate:

cd minggu-3-face-recognition/learning
python lesson-1/main.py

Want Quick Demo:

# Setup Week 6 first
cd minggu-6-attendance-system
python setup_week6.py  # Pilih [1] Copy dari Week 5

# Run attendance demo
cd learning/lesson-1
python main.py

---

Happy Learning! 🎉

Remember: Everyone starts from zero. Keep learning, don't give up!

---

Last Updated: December 13, 2025
Version: 3.0
License: Educational purposes