This project implements a face emotion detection system using YOLOv11, trained on a custom dataset to classify emotions into five distinct classes. The model utilizes the Ultralytics YOLO framework for real-time inference.

• Features
• Installation
• Usage
• License
• Contributing

• Real-time face emotion detection.
• Custom-trained YOLOv11 model for improved accuracy.
• Supports detection of five emotion classes.

To get started, clone the repository and install the required packages.

git clone https://github.com/alihassanml/Yolo11-Face-Emotion-Detection.git
cd Yolo11-Face-Emotion-Detection
pip install -r requirements.txt

Make sure you have the following dependencies installed:

• Python 3.x
• OpenCV
• Ultralytics YOLO

To run the model for face emotion detection, use the following script:

from ultralytics import YOLO
import cv2

# Load the trained model
model = YOLO('best.onnx') 

cap = cv2.VideoCapture(0)

while True:
    ret, frame = cap.read()
    if not ret:
        print("Failed to grab frame.")
        break  

    # Convert the frame to grayscale
    gray_image = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    
    # Convert grayscale to 3-channel image
    gray_image_3d = cv2.merge([gray_image, gray_image, gray_image]) 
    
    # Perform inference
    results = model(gray_image_3d)
    result = results[0]

    # Plot results
    try:
        annotated_frame = result.plot()
    except AttributeError:
        print("Error: plot() method not available for results.")
        break
    
    # Display the output
    cv2.imshow('YOLO Inference', annotated_frame)
    
    if cv2.waitKey(1) == 27:  # ESC key to exit
        break

cap.release()
cv2.destroyAllWindows()

The results are visualized using OpenCV, displaying the detected emotions in real-time.

This project is licensed under the MIT License. See the LICENSE file for details.

Contributions are welcome! Please feel free to submit a pull request or open an issue for discussion.

Feel free to edit the sections to match your project details better. You can also include screenshots or visualizations in the README for added clarity and engagement!