🌟 Count-Face-Video 🎥😊

Welcome to Count-Face-Video! 🚀 This project brings the magic of Facial Recognition into action, identifying and tracking faces in videos with the power of PCA & KNN. Let's dive into AI & Computer Vision with a bright smile! 😃✨

---

🎯 Overview

The AI-powered system detects, identifies, and counts faces in videos. Leveraging Principal Component Analysis (PCA) and K-Nearest Neighbors (KNN), this project processes video files to track face occurrences over time. 🤖📊

---

🔄 Project Workflow 📌

1️⃣ Training Phase 📚: Process dataset images, apply PCA and train KNN classifier.

2️⃣ Detection Phase 🎥: Read video frames, detect faces, classify identities.

3️⃣ Result Analysis 📊: Save detection results in output.txt with timestamps.

---

📂 Project Structure 🏗️

📁 Count-Face-Video/
├── 📂 dataset/                             # Dataset containing subfolders for each person
├── 📄 haarcascade_frontalface_default.xml  # Haar Cascade model for face detection
├── 📄 face_recognition_model.pkl           # Trained PCA + KNN model
├── 📓 Train and Detect.ipynb               # Jupyter Notebook for training & detection
├── 🎥 drstrange_.mp4                       # Video file for face detection
├── 📄 output.txt                           # Detection results with timestamps

---

✨ Features

✅ Detect & recognize faces using Haar Cascade + PCA + KNN.

✅ Process video files and track face occurrences over time.

✅ Log results with timestamps to monitor face appearances.

✅ Lightweight and easy to use! ⚡

---

🛠 Technologies & Libraries

🔹 Python 🐍
🔹 OpenCV 👀
🔹 NumPy 🔢
🔹 Scikit-learn 🧠
🔹 Pickle 📦

---

🚀 Installation

Ensure you have Python installed, then install dependencies:

pip install opencv-python numpy scikit-learn pickle-mixin

---

🎬 Usage

🏗 Training the Model and Running Face Detection on a Video 🎥

Run the Jupyter Notebook to train the facial recognition model:

jupyter notebook Train and Detect.ipynb

---

📊 Result and Findings

The detection results are stored in output.txt, showing occurrences with timestamps:

ben detected 560 times
Appears in time:
0 minute 0 second
0 minute 1 second
...
eli detected 39 times
Appears in time:
0 minute 0 second
0 minute 1 second
...

---

🚀 Future Improvements 🔮

✨ Implement deep learning-based recognition (e.g., CNN).

✨ Improve accuracy using additional feature extraction techniques.

✨ Add real-time video streaming support.

---

🤝 Contributing 🌍

Contributions are welcome! If you'd like to improve the project:

1️⃣ Fork the repository 🍴

2️⃣ Create a new branch 🌱

3️⃣ Make your improvements ✨

4️⃣ Submit a pull request 🔄

🌟 If you love this project, don't forget to star ⭐ the repository and contribute! 🙌

---

💖 Happy Coding & Keep Innovating! 💖 ✨