Face synthesis with generative adversarial networks

Zhengqiao Li; Tianjin Liu; Xinyuan Wei; Letian Zhou

doi:10.1117/12.2639259

17 May 2022 Face synthesis with generative adversarial networks

Zhengqiao Li, Tianjin Liu, Xinyuan Wei, Letian Zhou

Proceedings Volume 12259, 2nd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2022); 122595I (2022) https://doi.org/10.1117/12.2639259
Event: 2nd International Conference on Applied Mathematics, Modelling, and Intelligent Computing, 2022, Kunming, China

Abstract

Nowadays, machine learning has become a trend in computer programming techniques. It is a learning method that builds a model based on training data, in order to make predictions or decisions without being explicitly programmed to do so. Generative adversarial network (GAN) is the most advanced technology in unsupervised learning. The GAN is composed of a generator and a discriminator. Generator aims to generate realistic samples and discriminators to identify real data and generated fake data. GAN has been successfully implemented for solving a variety of tasks such as face synthesis, facial attribute manipulation, and image to image translation. Face synthesis gains great attention recently, especially after the emerge of Deepfakes. With this technology, realistic images or media can be easily created. In this paper, we will start with the introduction of GAN and then summarize and analyze five popular methods for face synthesis using GAN technology. Finally, we discuss some problems of existing methods and point out a future direction for face synthesis.

Citation Download Citation

Zhengqiao Li, Tianjin Liu, Xinyuan Wei, and Letian Zhou "Face synthesis with generative adversarial networks", Proc. SPIE 12259, 2nd International Conference on Applied Mathematics, Modelling, and Intelligent Computing (CAMMIC 2022), 122595I (17 May 2022); https://doi.org/10.1117/12.2639259

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available