Subscription

Explore Products

Best Sellers

New Releases

Books

Videos

Audiobooks

Learning Hub

Conferences

Free Learning

You're reading from Hands-On Generative Adversarial Networks with PyTorch 1.x Implement next-generation neural networks to build powerful GAN models using Python

Product type Paperback

Published in Dec 2019

Publisher Packt

ISBN-13 9781789530513

Length 312 pages

Edition 1st Edition

Languages

Python

Tools

PyTorch

Concepts

Data Science

Authors (2):

John Hany

Greg Walters

View More author details

Table of Contents (15) Chapters

Preface

1. Section 1: Introduction to GANs and PyTorch FREE CHAPTER

2. Generative Adversarial Networks Fundamentals

3. Getting Started with PyTorch 1.3

4. Best Practices for Model Design and Training

5. Section 2: Typical GAN Models for Image Synthesis

6. Building Your First GAN with PyTorch

7. Generating Images Based on Label Information

8. Image-to-Image Translation and Its Applications

9. Image Restoration with GANs

10. Training Your GANs to Break Different Models

11. Image Generation from Description Text

12. Sequence Synthesis with GANs

13. Reconstructing 3D models with GANs

14. Other Books You May Enjoy

Leave a review - let other readers know what you think

Generative image inpainting

We know that GANs, if trained properly, are capable of learning the latent distribution of data and using that information to create new samples. This extraordinary ability of GANs makes them perfect for applications such as image inpainting, which is filling the missing part in images with plausible pixels.

In this section, we will learn how to train a GAN model to perform image inpainting, based on the work of Jiahui Yu, Zhe Lin, Jimei Yang, et. al. in their paper, Generative Image Inpainting with Contextual Attention. Although an updated version of their project has been published (http://jiahuiyu.com/deepfill2), the source code is not yet open source at the time of writing. Therefore, we should try to implement the model in PyTorch based on the source code of its previous version for TensorFlow (https://github.com/JiahuiYu/generative_inpainting...

The rest of the chapter is locked

A free Packt account unlocks extra newsletters, articles, discounted offers, and much more. Start advancing your knowledge today.

Unlock this book and the full library FREE for 7 days

Get unlimited access to 7000+ expert-authored eBooks and videos courses covering every tech area you can think of

Start free trial

Renews at €18.99/month. Cancel anytime

Authors (2)

John Hany

John Hany received his master's degree and bachelor's degree in calculational mathematics at the University of Electronic Science and Technology of China. He majors in pattern recognition and has years of experience in machine learning and computer vision. He has taken part in several practical projects, including intelligent transport systems and facial recognition systems. His current research interests lie in reducing the computation costs of deep neural networks while improving their performance on image classification and detection tasks. He is enthusiastic about open source projects and has contributed to many of them.

See other products by John Hany

Greg Walters

Greg Walters has been involved with computers and computer programming since 1972. He is well-versed in Visual Basic, Visual Basic .NET, Python, and SQL and is an accomplished user of MySQL, SQLite, Microsoft SQL Server, Oracle, C++, Delphi, Modula-2, Pascal, C, 80x86 Assembler, COBOL, and Fortran. He is a programming trainer and has trained numerous people on many pieces of computer software, including MySQL, Open Database Connectivity, Quattro Pro, Corel Draw!, Paradox, Microsoft Word, Excel, DOS, Windows 3.11, Windows for Workgroups, Windows 95, Windows NT, Windows 2000, Windows XP, and Linux. He is semi-retired and has written over 100 articles for Full Circle Magazine. He is also a musician and loves to cook. He is open to working as a freelancer on various projects.

See other products by Greg Walters