This repository contains a project aimed at detecting human emotions in real-time using the YOLOv8 model. The model is trained to classify four distinct emotions: angry, happy, sad, and surprised.
Emotion recognition is a critical task in various fields like mental health, security, and human-computer interaction. This project leverages the YOLOv8 model, a state-of-the-art object detection algorithm, to classify emotions from images or video streams.
The YOLOv8 model has been trained to detect the following emotions:
- Angry
- Happy
- Sad
- Surprised
We used the YOLOv8 architecture, which offers superior speed and accuracy for real-time object detection. The model was fine-tuned to detect facial features and classify them into the aforementioned emotions.
The model was trained on a custom dataset containing images annotated with facial expressions representing different emotions. The dataset includes a diverse set of images representing the following classes:
- Angry
- Happy
- Sad
- Surprised
The dataset is split into training, validation, and test sets to ensure the model generalizes well to unseen data.
The YOLOv8 model was trained using the following key parameters:
- Batch size: 16
- Image size: 640x640
- Epochs: 100
- Learning rate: 0.001
The model achieved high accuracy in detecting emotions across various test images, demonstrating the effectiveness of YOLOv8 for this task.
- Clone this repository:
git clone https://github.com/Musawer1214/Emotion-Detection.git cd Emotion-Detection-yolov8
