Jun 2024 – Jun 2024AI & Data

Facial Expression Recognition with PyTorch

A machine learning project designed to accurately classify human facial expressions from images using a trained PyTorch model.

Rationale

Why Facial Expression Recognition? Associated with Auburn University at Montgomery, this project was an exciting deep dive into machine learning to classify human emotions from raw pixels. It highlights my proficiency in machine learning, deep learning, and computer vision architectures.

Tech Stack

PyTorchPythonDeep LearningCNN

Key Highlights

▹Preprocessing of facial images for robust feature extraction.
▹Training a convolutional neural network (CNN) for precise expression recognition.
▹Achieving high accuracy in classifying diverse expressions such as happiness, sadness, anger, and surprise.

Architecture Details

1. Data Preprocessing

Cleaning and manipulating facial images to ensure proper alignment and normalize data distributions.

2. Deep Learning Modeling

Built entirely within PyTorch, implementing a deep Convolutional Neural Network (CNN).
Multiple layers engineered to capture nuanced edge and texture features defining distinct human expressions mathematically.