Skip to content

Latest commit

 

History

History
25 lines (20 loc) · 955 Bytes

README.md

File metadata and controls

25 lines (20 loc) · 955 Bytes

Recognition of Emotion from Images Using Deep Learning

Abstract

This project explores human emotion recognition from images using deep learning, focusing on CNN architectures including a baseline CNN, a modified VGGNet, and a pre-trained ResNet-50 with transfer learning. The FER-2013 dataset, containing grayscale facial expression images, was used for training and evaluation.

Installation

  • Flask==3.0.0
  • gunicorn==20.1.0
  • tensorflow==2.8.0
  • numpy==1.22.3
  • Pillow==9.0.1
  • matplotlib==3.5.1
  • seaborn==0.11.2
  • Werkzeug==3.0.0

Dataset

FER-2013: Contains 35887 grayscale images of faces across seven emotion classes (0=Angry, 1=Disgust, 2=Fear, 3=Happy, 4=Sad, 5=Surprise, 6=Neutral)

Models

  1. Baseline Model: Custom CNN
  2. VGGNet Modified CNN
  3. ResNet-50 with Transfer Learning

Results

The models were evaluated based on accuracy, loss, and confusion matrices, with ResNet-50 showing superior performance (71% Test Accuracy)