https://github.com/mktechai-0786/facial-emotion-detector

Face Emotion Recognition using FER Lib
https://github.com/mktechai-0786/facial-emotion-detector

deep-learning fer2013 machine-learning mtcnn-face-detection opencv python tenserflow

Last synced: 21 days ago
JSON representation

Face Emotion Recognition using FER Lib

• Host: GitHub
• URL: https://github.com/mktechai-0786/facial-emotion-detector
• Owner: MKTechAI-0786
• Created: 2024-12-16T16:55:06.000Z (4 months ago)
• Default Branch: main
• Last Pushed: 2024-12-16T17:10:51.000Z (4 months ago)
• Last Synced: 2025-04-09T00:07:15.237Z (21 days ago)
• Topics: deep-learning, fer2013, machine-learning, mtcnn-face-detection, opencv, python, tenserflow
• Language: Python
• Homepage:
• Size: 4.88 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Facial-Emotion-Detector

This project is a Facial Emotion Recognition System that uses a webcam or video input to detect faces and identify their emotions in real-time. The program utilizes the FER (Facial Expression Recognition) library for emotion classification and OpenCV for video capture and display.

#Features

  Real-time emotion detection using a webcam.

  Supports multiple emotion classes:

    Happy, Sad, Angry, Neutral, Fear, Disgust, Surprise

  Draws bounding boxes around detected faces and displays the most likely emotion.

#Accuracy

  Controlled environments: ~70-80% accuracy (good lighting, frontal faces).

  Real-world scenarios: ~50-60% accuracy (varied lighting and non-frontal faces).

  Per-emotion classification accuracy (approximate):

    Happy: ~93%

    Neutral: ~60%

    Sad: ~75%

    Angry: ~70%

    Surprise: ~60%

    Fear, Disgust: ~58%

#Software Requirements

Make sure the following software and libraries are installed before running the project:

  Python 3.8 or later

  Required Python packages (install using pip):

    "pip install opencv-python-headless fer"

  OpenCV: For video capture and processing.

  FER: A library for facial expression recognition.

#Installation

  1. Clone the Repository

    git clone https://github.com/yourusername/facial-emotion-recognition.git

    cd facial-emotion-recognition

  2. Install Dependencies

     Install all required libraries using pip:

      pip install -r requirements.txt

  3. Run the Program

    Run the script to start emotion detection via webcam:

      python emotion_detector.py

     

5. Exit the Program

    Press the q key to exit the webcam feed.

#Procedure

  1. The program captures live video feed from the webcam.

  2. It detects faces in real-time using the FER library.

  3. For each detected face:

    3.1 A bounding box is drawn around the face.

    3.2 The most probable emotion (e.g., happy, sad, angry, etc.) is displayed above the box.

  4. Results are continuously displayed in a video window.

  5. The program exits when the q key is pressed.

#Conclusion

  This project provides a basic framework for detecting facial emotions using pre-trained models in the FER library. While it works well in controlled environments, accuracy may decrease in real-world settings due to lighting, occlusions, and non-frontal faces. For improved performance, consider using advanced models such as DeepFace or custom fine-tuned neural networks.

#Future Improvements

  Integrate more advanced models like DeepFace for higher accuracy.

  Add support for analyzing emotion trends over time.

  Improve robustness under varying lighting conditions.