O'Reilly logo
live online training icon Live Online training

Deep Learning for Machine Vision: Complete Artificial Intelligence Series

Visual Processing and Creativity with Convolutional and Generative Adversarial Networks

Jon Krohn

Relatively obscure a few short years ago, Deep Learning is ubiquitous today across data-driven applications as diverse as machine vision, super-human game-playing, and natural language processing (NLP). This Live Training builds on the fundamentals of Deep Learning to develop a specialization in Machine Vision, a foundation of contemporary Artificial Intelligence.

Since a Deep Learning algorithm named AlexNet smashed visual-recognition benchmarks in 2012, machine-vision applications have led the charge in the Deep Learning wave that has since engulfed the world. In this training, we will detail how the Convolutional Neural Networks (CNNs) like AlexNet that predominate contemporary Machine Vision function. We’ll then employ CNNs ourselves to attain remarkably accurate results in object detection and image segmentation tasks. To round the training off, we’ll build a revolutionary Generative Adversarial Network (GAN) to further leverage CNNs and produce convincing synthetic images!

To facilitate an intuitive understanding of Machine Vision, essential theory will be introduced visually and pragmatically. Theory will immediately be brought to life with interactive demos and hands-on exercises featuring Keras, the wildly-popular Python API for TensorFlow 2.0.

This is part of Jon Krohn’s Complete Artificial Intelligence Series, a collection of interactive trainings that together comprehensively cover the foundations of modern AI approaches. The recommended progression through the Series is to take one of these two introductory sessions:

Following either of the introductory sessions (or if you’re familiar with the content covered in Chapters 1 and 5-9 of Jon Krohn’s Deep Learning Illustrated book), you’re well-prepared to specialize in any of the other Live Trainings in the Complete Artificial Intelligence Series, which can be undertaken in any order you fancy:

What you'll learn-and how you can apply it

  • Understand the high-level theory and key language around Convolutional Neural Networks, a foundational technique within contemporary Artificial Intelligence systems
  • Build a Convolutional Neural Network for state-of-the-art performance in object recognition and image segmentation
  • Architect a sophisticated Generative Adversarial Network to enable machines to create high-quality, synthetic images

This training course is for you because...

  • You already have a working understanding of the fundamentals of Deep Learning
  • You want to design Convolutional Neural Networks for use in Machine Vision applications
  • You’re curious as to how a Generative Adversarial Network can make art all on its own

Prerequisites

  • Experience with an object-oriented programming language, e.g., Python (all code demos during the training will be in Python)
  • A working understanding of the fundamentals of Deep Learning would make it a lot easier to follow along with the training

Materials, downloads, or Supplemental Content needed in advance:

  • During class, we’ll work on Colab notebooks interactively in the cloud

Resources:

About your instructor

  • Jon Krohn is Chief Data Scientist at the machine learning company untapt. He is the presenter of a popular series of tutorials on artificial neural networks, including Deep Learning with TensorFlow, and is the author of Deep Learning Illustrated, the acclaimed book released by Pearson in 2019. Jon holds a doctorate in neuroscience from Oxford University and has been publishing on machine learning in leading academic journals since 2010. He teaches his deep learning curriculum at the NYC Data Science Academy as well as Columbia University. Along with researchers at Columbia’s medical center, Dr. Krohn holds a National Institutes of Health grant to automate medical image processing with deep learning.

Schedule

The timeframes are only estimates and may vary according to how the class is progressing

Segment 1: Introducing Deep Learning for Machine Vision (60 min)

  • Training Overview
  • Machine Vision Applications
  • Review of Prerequisite Deep Learning Theory
  • Essential Theory of Convolutional Neural Networks

Segment 2: Convolutional Neural Networks in Practice with Keras (90 min)

  • LeNet-5
  • AlexNet
  • VGGNet
  • Residual Networks (ResNet)
  • U-Net
  • Image Classification
  • Object Detection
  • Semantic Image Segmentation
  • Transfer Learning

Segment 3: Generative Adversarial Networks (30 min)

  • “A Boozy All-Nighter”: How GANs were Born
  • Essential GAN Theory
  • A Cartoon-Drawing GAN in Keras