👁️ Object, Eye & Emotion (Smile) Detection using YOLO and OpenCV

This project uses YOLO (You Only Look Once) and OpenCV to perform real-time object detection, eye tracking, and emotion/smile detection through a webcam or video input.

It demonstrates how modern deep learning models can be combined with traditional computer vision techniques to analyze faces and emotions effectively.

🧠 Project Overview

🧍‍♂️ Object Detection — Detects multiple classes such as people, cars, and objects using YOLO.

👁️ Eye Detection — Uses Haar cascades or deep learning for accurate eye localization.

😊 Emotion / Smile Detection — Detects facial expressions and identifies smiles.

📷 Real-Time Processing — Works with live webcam feed or pre-recorded videos.

🖼️ Example Outputs 🔹 Object Detection

🔹 Eye Detection

🔹 Emotion / Smile Detection

🧰 Tech Stack Component Technology Language Python 3.x Frameworks OpenCV, NumPy Model YOLOv8 / YOLOv5 Pre-trained Models Haar Cascades / CNN for Smile & Eye Detection Visualization Matplotlib / OpenCV ⚙️ Installation & Setup 1️⃣ Clone the Repository git clone https://github.com/risingstar185/yolo_haarcascade.git

3️⃣ Install Dependencies pip install -r requirements.txt

4️⃣ Download YOLO Weights

Download YOLO pre-trained weights from:

YOLOv8 Official

or use older YOLOv3 weights:

wget https://pjreddie.com/media/files/yolov3.weights

Place them in the project directory (weights/ folder).

▶️ Usage 🔹 For Object Detection python object_detection.py --source 0 # For webcam python object_detection.py --source path/to/video.mp4

🔹 For Eye & Smile Detection python face_eye_smile_detection.py

🔹 For Emotion Detection (Optional CNN Model) python emotion_detection.py

🧪 Sample Results Detection Type Example Output Object ✅ Person, Car, Laptop Eye 👁️ Both eyes detected Smile 😄 Smile recognized 💡 Future Improvements

🔊 Add voice feedback for detection results.

🌈 Integrate with a dashboard (Flask/Streamlit).

🧍‍♀️ Add full facial emotion recognition (7+ categories).

☁️ Deploy as a web app using FastAPI or Streamlit Cloud.

📜 License

This project is licensed under the MIT License — you are free to use, modify, and distribute it with attribution.

🙌 Author

Your Name :Ayush Patel