Convolutional Neural Networks and Image Processing II (EN)

Description

This is a continuation of the Convolutional Networks and Image Processing course, in which we will focus in more detail on data preprocessing and advanced deep learning techniques for image processing.

In addition to classifications well known from the previous course, we will focus on segmentation, object detection and especially advanced applications of generative adversarial networks (GAN) such as resolution enhancement, noise removal and Deep fake generation.

Contents

• Architectures of neural networks for image processing (convolution, deconvolution, pooling, residual connection)
• Large neural networks for image processing (VGG 16 and ResNet)
• Image segmentation (U-net, object detection)
• A practical example of image segmentation
• Generative Adversarial Networks (GANs)
• A practical example for image generation
• Superresolution (Upsampling, a practical example of increasing image resolution using GAN)
• A practical larger-scale project to predict real estate prices using a combination of tabular and image data

Prerequisites

• Basic knowledge of Python programming
• High School Mathematics
• Course-level knowledge of machine learning Introduction to Machine Learning
• Course-level knowledge Convolutional neural networks and image processing