From Autoencoders and Adversarial Networks to Deepfakes

Table of Contents
About the Author
About the Technical Reviewer
Acknowledgments
Introduction
Chapter 1: The Basics of Deep Learning
	Prerequisites
	The Perceptron
	The Multilayer Perceptron
		Backpropagation
		Stochastic Gradient Descent
	PyTorch and Deep Learning
	Understanding Regression
		Over- and Underfitting
	Classifying Classes
		One-Hot Encoding
		Classifying MNIST Digits
	Conclusion
Chapter 2: Unleashing Generative Modeling
	Unsupervised Learning with Autoencoders
	Extracting Features with Convolution
	The Convolutional Autoencoder
	Generative Adversarial Networks
	Deep Convolutional GAN
	Conclusion
Chapter 3: Exploring the Latent Space
	Understanding What Deep Learning Learns
		Deep Learning Function Approximation
		The Limitations of Calculus
		Deep Learning Hill Climbing
		Over- and Underfitting
	Building a Variational Autoencoder
	Learning Distributions with the VAE
	Variability and Exploring the Latent Space
	Conclusion
Chapter 4: GANs, GANs, and More GANs
	Feature Understanding and the DCGAN
	Unrolling the Math of GANs
	Resolving Distance with WGAN
	Discretizing Boundary-Seeking GANs
	Relativity and the Relativistic GAN
	Conditioning with CGAN
	Conclusion
Chapter 5: Image to Image Content Generation
	Segmenting Images with a UNet
		Uncovering the Details of a UNet
	Translating Images with Pix2Pix
	Seeing Double with the DualGAN
	Riding the Latent Space on the BicycleGAN
	Discovering Domains with the DiscoGAN
	Conclusion
Chapter 6: Residual Network GANs
	Understanding Residual Networks
	Cycling Again with CycleGAN
	Creating Faces with StarGAN
	Using the Best with Transfer Learning
	Increasing Resolution with SRGAN
	Conclusion
Chapter 7: Attention Is All We Need!
	What Is Attention?
		Understanding the Types of Attention
		Applying Attention
	Augmenting Convolution with Attention
	Lipschitz Continuity in GANs
		What Is Lipschitz Continuity?
	Building the Self-Attention GAN
	Improving on the SAGAN
	Conclusion
Chapter 8: Advanced Generators
	Progressively Growing GANs
	Styling with StyleGAN Version 2
		Mapping Networks
		Style Modules
			Removing Stochastic/Traditional Input
			Stochastic Variation (Noisy Inputs)
			Mixing Styles
			Truncation of W
			Hyperparameter Tuning
		Frechet Inception Distance
		StyleGAN2
			Weight Demodulation
			Path Length Regularization
			Lazy Regularization
			No Growing
			Large Networks
	DeOldify and the New NoGAN
		Colorizing and Enhancing Video
	Being Artistic with ArtLine
	Conclusion
Chapter 9: Deepfakes and Face Swapping
	Introducing the Tools for Face Swapping
	Gathering the Swapping Data
		Downloading YouTube Videos for Deepfakes
	Understanding the Deepfakes Workflow
		Extracting Faces
		Sorting and Trimming Faces
		Realigning the Alignments File
	Training a Face Swapping Model
	Creating a Deepfake Video
		Encoding the Video
	Conclusion
Chapter 10: Cracking Deepfakes
	Understanding Face Manipulation Methods
	Techniques for Cracking Fakes
		Handcrafted Features
		Learning-Based Features
		Artifacts
	Identifying Fakes in Deepfakes
	Conclusion
Appendix A:
Running Google Colab Locally
Appendix B:
Opening a Notebook
Appendix C:
Connecting Google Drive and Saving
Index