🎥 Emotional State Detection Through Video

This project implements an emotional state detection system that analyzes the emotions of a person from a video file. The system uses deep learning models to detect facial expressions from video frames and predict the corresponding emotional state, all while creating a seamless and interactive user experience. 🌟

Project Structure

app/
│
├── .dockerignore                # Specifies files to be excluded from Docker builds
├── .gitattributes               # Git settings for handling line endings and other attributes
├── Dockerfile                   # Dockerfile for containerizing the application
├── main.py                      # Main application script that runs the emotion detection pipeline
├── requirements.txt             # Required Python packages for the project
├── sample.py                    # Example script for reference
│
├── extracted_data/
│   └── emotions_data.csv         # Stores the detected emotions from video processing
│
├── initial_code/
│   └── Emotion_detector.py       # Initial code or reference implementation for emotion detection
│
├── output/
│   └── output_video.mp4          # Output video with detected emotions overlaid on faces
│
├── pages/
│   └── Emotion_Prediction.py     # Additional pages for the Streamlit app (if needed)
│
└── utils/
    ├── haarcascade_frontalface_default.xml  # Haar Cascade file for face detection
    ├── Model.h5                            # Pre-trained emotion detection model
    └── temp_video.mp4                      # Temporary storage for uploaded video

1. Overview

📊 The application leverages a pre-trained Convolutional Neural Network (CNN) model to predict emotions based on facial expressions detected in a video. Using Streamlit for a user-friendly interface, and OpenCV to process video frames and detect faces, the application predicts emotions, stores the data in a CSV file, and overlays labels on the processed video.

Additionally, the processed video can be previewed with real-time emotion annotations for each face in the video. This adds an engaging touch by showing emotional transitions as the video plays! 🎬😃

2. Features

✨ Key Highlights:

• 🎥 Video Upload: Users can upload an .mp4 video for emotion detection.
• 👤 Face Detection: Uses OpenCV's Haar Cascade to detect faces in each frame.
• 🤖 Emotion Prediction: Predicts emotions using a pre-trained model (Model.h5) and labels the dominant emotion for each detected face.
• 📊 CSV Output: Extracted emotions are saved in a CSV file for further analysis.
• 📽️ Video Output: Generates an output video with detected emotions overlaid on each detected face.
• 🖥️ Streamlit Interface: Simple, interactive web interface for video upload, processing, and result preview.

3. Emotion Classes

The model is trained to detect the following emotions:

• 😡 Angry
• 😒 Disgust
• 😱 Fear
• 😊 Happy
• 😢 Sad
• 😲 Surprise
• 😐 Neutral

These emotions are visually displayed on the detected faces throughout the video, making it easier to observe how emotions change over time. ⏳

4. How It Works

1. 📥 Model Loading: The pre-trained CNN model (Model.h5) is loaded along with the Haar Cascade classifier for face detection.
2. 🎦 Video Processing: The user uploads a video file, and the application processes it frame by frame.
3. 🧠 Face Detection: Faces are detected in each frame using the Haar Cascade classifier.
4. 🤔 Emotion Prediction: For each detected face, the emotion is predicted by the CNN model, and the dominant emotion is labeled.
5. 🎬 Output Generation: The processed video with emotion labels is displayed to the user, and the emotion data is saved in emotions_data.csv. The video playback also reflects changes in emotional states throughout the video! 📊🎥

5. Installation

To run the project locally, follow these steps:

1. Clone the repository:

   git clone https://github.com/itzmechandruganeshan/Emotion_Prediction.git

2. Navigate to the project directory:

   cd app

3. Install the required packages:

   pip install -r requirements.txt

4. Run the Streamlit app:

   streamlit run main.py

6. Dependencies

The project requires the following Python packages:

• opencv-python 🖼️
• tensorflow 🤖
• keras 🔬
• numpy 🔢
• pandas 📊
• streamlit 🌐

These can be installed using the requirements.txt file.

7. Usage

1. 🚀 Launch the application using Streamlit.
2. 📤 Upload a video file in .mp4 format.
3. 🕰️ Wait for the video processing to complete.
4. 🎥 The detected emotions will be overlaid on the faces in the video.
5. 💾 Download the generated CSV file with emotion data for further analysis.