Facial Emotion Recognition (FER)

Project Description

Facial Emotion Recognition (FER) is a project that detects multiple faces, draws bounding boxes around each, and determines the emotion of each detected face. The system uses a Haarcascade pre-trained model for face detection and a Convolutional Neural Network (ConvNN) trained on a dataset for emotion classification. The application operates in real-time.

Face Detection

• Haarcascade Pre-trained Model: Link to model

Emotion Classification

• Convolutional Neural Network (ConvNN) trained on the FER2013 dataset: Link to dataset

Manual

For Inference

python EmoRec.py

For Training

Check the notebook main.py

Test Video

Watch the demo video

Installation

To install the required dependencies, use the following command:

pip install -r requirements.txt

Requirements

• Python 3.x
• OpenCV
• TensorFlow
• Numpy
• ...

Results

Include any relevant results, performance metrics, or accuracy scores from the FER system.

---

Feel free to customize the sections, add specific dependencies to the requirements.txt file, and include any other information that might be relevant for users and contributors.