5: Visual Computing I

Date: 22nd November 2023

💡 Visual computing includes various techniques to process and analyse visual data, such as images and videos. Image representation involves encoding these into digital formats, commonly using pixels with RGB values. Preprocessing techniques like normalization, resizing, and rotation can help us produce a more suitable dataset for training neural networks. Convolutional Neural Networks (CNNs) play a huge role in allowing models to understand images, using kernels for feature extraction. We will also explore classic CNN architectures like AlexNet, VGG, and ResNet for image classification and object detection, showing how these architectural designs have revolutionized the field of computer vision.💡

You can access our demonstration notebook here: 📘 Tutorial 5 Notebook

You can access our slides here: 💻 Tutorial 5 Slides

The recording from this session is available here: 🎤 Tutorial 5 Recording