Emotion Detection with OpenCV and Keras

Overview

This project utilizes OpenCV and Keras to perform real-time emotion detection on faces in a video stream or file. The system uses a pre-trained deep learning model for emotion classification. Detected emotions include Angry, Disgusted, Fearful, Happy, Neutral, Sad, and Surprised.

Dependencies

cv2: OpenCV library for computer vision tasks. numpy: NumPy library for numerical operations. Keras: Deep learning library for building and training models.

Setup

Clone the repository: git clone https://github.com/sanj16/emotion-detection.git

Install dependencies: pip install opencv-python numpy keras

Download the pre-trained model files: emotion_model.json emotion_model.h5

Run the script: python emotion_detection.py

Usage

The script can be configured to use either the default camera or a specified video file. Press 'q' to exit the application.

License