Emotion Detection with Real-time Camera Access

Overview

This project detects human emotions in real-time using a Convolutional Neural Network (CNN) trained on the FER-2013 dataset. The model classifies facial expressions into seven different categories and processes real-time input from a webcam using OpenCV.

---

Dataset Structure

The FER-2013 dataset contains labeled images of facial expressions divided into training and testing sets.

dataset/
│── train/
│   ├── happy/
│   ├── anger/
│   ├── fear/
│   ├── surprise/
│   ├── neutral/
│   ├── disgust/
│   ├── sad/
│── test/
│   ├── happy/
│   ├── anger/
│   ├── fear/
│   ├── surprise/
│   ├── neutral/
│   ├── disgust/
│   ├── sad/

Each folder contains thousands of images categorized based on emotion labels.

---

Project Structure

emotiondetection/
│── dataset/               # FER-2013 dataset
│── models/                # Saved trained model
│── src/
│   ├── train_model.py      # Model training script
│   ├── test_model.py       # Model testing script
│   ├── realtime_emotion.py # Real-time emotion detection
│── requirements.txt        # Required dependencies
│── README.md               # Project documentation

---

Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/your-username/emotion-detection-ai.git
cd emotion-detection-ai

2️⃣ Install Dependencies

Ensure you have Python installed, then run:

pip install -r requirements.txt

---

How to Run?

🔹 Train the Model

python src/train_model.py

• Trains the CNN model on the FER-2013 dataset.
• Saves the trained model in the models/ directory.

🔹 Test the Model

python src/test_model.py

• Evaluates the trained model on the test dataset.
• Outputs accuracy, loss, and sample predictions.

🔹 Run Real-time Emotion Detection

python src/realtime_emotion.py

• Accesses the webcam using OpenCV.
• Detects and classifies facial expressions in real time.

---

Features

✅ Trained on FER-2013 dataset (7 emotion classes: Happy, Sad, Anger, Fear, Disgust, Neutral, Surprise).
✅ Real-time emotion detection using OpenCV.
✅ CNN-based deep learning model with high accuracy.
✅ Fast & lightweight implementation with optimized performance.

---

Technologies Used

• Python
• TensorFlow / Keras
• OpenCV
• NumPy & Pandas
• Matplotlib (for visualization)

---

License

This project is licensed under the MIT License.

---

Contributing

Feel free to contribute by opening issues or pull requests. If you find this project helpful, give it a ⭐ on GitHub!

🚀 Happy Coding! 😊