Emotion Detector Project

Overview

This project uses a deep learning model (CNN) to detect emotions from images. The model is trained on datasets like FER2013 and integrates with a Flask API for emotion detection. Additionally, a Vue 3 web interface is developed for user interaction.

---

Features

• Emotion Detection: Detects emotions such as Angry, Happy, Sad, etc., from facial images.
• Real-time Webcam Support: Enables real-time emotion detection using a webcam.
• Web Interface: Provides an intuitive frontend built with Vue 3.
• API Integration: Backend powered by Flask, connecting the model and the web interface.

---

Implementation Steps

Step 1: Data Preparation

• Datasets: Use datasets like FER2013, CK+, or JAFFE for training.
• Preprocessing:
  • Resize images to 48x48 pixels (grayscale).
  • Normalize pixel values to range [0, 1].

---

Step 2: Build the Model

• Architecture: Build a Convolutional Neural Network (CNN) with:
  • Convolutional layers
  • MaxPooling layers
  • Fully connected layers
• Activation Functions: Use ReLU for hidden layers and Softmax for the output layer.

---

Step 3: Train the Model

• Data Splits: Divide the dataset into training, validation, and test sets.
• Training Framework: Use TensorFlow/Keras.
• Evaluation Metrics: Track metrics like accuracy, precision, recall, and confusion matrix.

---

Step 4: Deploy API

• Framework: Use Flask to build a RESTful API.
• Functionality:
  • Accept uploaded images.
  • Process images and return emotion predictions.
• Model Integration: Load the trained model within the Flask app.

---

Step 5: Web Interface Development

Tools:

• Frontend Framework: Vue 3.
• Face Detection Library: face-api.js for detecting and cropping faces.

Steps:

1. Install Face API:

   npm install face-api.js