Foundations and Applications of Neural Networks in Image Classification: A PyTorch-Based Approach

Abstract

This paper discusses basic neural network concepts before looking at more specific cases in computer vision and image classification. This includes topics such as neural layers, backpropagation, convolutions, residuals, and two prominent image classification methods known as ResNet18 and Visual Transformers. Its goal is to provide a beginner-friendly introduction to the topic without previous backgrounds, and achieve this with proofs, media, and examples written with Python and its PyTorch library.