README.md

metadata

language: en
license: mit
tags:
  - keras
  - emotion-detection
  - facial-expression
  - image-classification
  - fer2012
model-index:
  - name: Facial Emotion Recognition (FER-2012)
    results:
      - task:
          type: image-classification
          name: Image Classification
        dataset:
          name: FER-2012
          type: fer2012
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.84

Facial Expression Recognition using CNN (FER-2012 Dataset)

This repository contains a Convolutional Neural Network (CNN) model trained using the FER-2012 dataset to classify facial expressions into seven emotion categories.

Model Details

• Framework: TensorFlow / Keras
• Input: 48x48 grayscale facial image
• Output: Emotion class (0–6)
• Model Format: .keras (Keras native format)

Emotion Classes

0 → Angry  
1 → Disgust  
2 → Fear  
3 → Happy  
4 → Sad  
5 → Surprise  
6 → Neutral