https://github.com/anidipta/task-5

Develop a real-time emotion detection system that operates on streaming video data and identifies the predominant emotion in each frame
https://github.com/anidipta/task-5

Last synced: 7 days ago
JSON representation

Develop a real-time emotion detection system that operates on streaming video data and identifies the predominant emotion in each frame

• Host: GitHub
• URL: https://github.com/anidipta/task-5
• Owner: Anidipta
• License: apache-2.0
• Created: 2024-04-23T15:03:55.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-08-19T16:36:17.000Z (5 months ago)
• Last Synced: 2024-08-19T19:39:34.800Z (5 months ago)
• Language: Jupyter Notebook
• Size: 10.1 MB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# 🧠 Task-5: Real-Time Emotion Detection System

### 🎯 Objective:

Develop a real-time emotion detection system that operates on streaming video data and identifies the predominant emotion in each frame.

### 📂 Project Files:

- **`README.md`** - Updated to include project details and instructions. 

- **`null_5 (1).ipynb`** - Jupyter notebook for emotion detection analysis.

- **`Task 5 Report.pdf`** - Detailed report of Task 5.

- **`null 5 app.py`** - Python script for the real-time emotion detection app.

- **`null 5 accuracy.png`** - Accuracy metrics of the emotion detection model.

- **`null 5 confusion matrix.png`** - Confusion matrix showing model performance.

- **`null 5 loss.png`** - Loss graph of the model training process.

- **`.gitignore`** - Git ignore file.

- **`LICENSE`** - Project license.

### 📸 Sample Demo Pictures:

Here are some sample images from the system in action:

![Sample Image 1](output/s1.jpg?raw=true)

![Sample Image 2](output/s2.jpg?raw=true)

![Sample Image 3](output/s3.jpg?raw=true)

### 📋 Project Highlights:

- **Real-Time Processing:** The system processes streaming video data to detect and classify emotions in real-time.

- **Emotion Identification:** Identifies predominant emotions in each frame of the video stream.

- **User-Friendly Interface:** Integrated into a Python application for easy use and deployment.

### 🚀 Development Details:

- **Technologies Used:** TensorFlow, Keras, OpenCV, NumPy, Pandas, Streamlit.

- **Models:** Utilized deep learning models for emotion recognition.

- **Performance Metrics:** Includes accuracy, confusion matrix, and loss graphs to evaluate the system's performance.

Feel free to explore the provided files and demo images to understand the system's capabilities and implementation!