Hyper-GAN: Transferring Unconditional to Conditional GANs with HyperNetworks

12/04/2021
by   Héctor Laria, et al.
0

Conditional GANs have matured in recent years and are able to generate high-quality realistic images. However, the computational resources and the training data required for the training of high-quality GANs are enormous, and the study of transfer learning of these models is therefore an urgent topic. In this paper, we explore the transfer from high-quality pre-trained unconditional GANs to conditional GANs. To this end, we propose hypernetwork-based adaptive weight modulation. In addition, we introduce a self-initialization procedure that does not require any real data to initialize the hypernetwork parameters. To further improve the sample efficiency of the knowledge transfer, we propose to use a self-supervised (contrastive) loss to improve the GAN discriminator. In extensive experiments, we validate the efficiency of the hypernetworks, self-initialization and contrastive loss for knowledge transfer on several standard benchmarks.

READ FULL TEXT

page 7

page 8

page 13

page 14

research
07/30/2023

Stylized Projected GAN: A Novel Architecture for Fast and Realistic Image Generation

Generative Adversarial Networks are used for generating the data using a...
research
10/27/2018

Self-Supervised GAN to Counter Forgetting

GANs involve training two networks in an adversarial game, where each ne...
research
12/08/2020

Data Instance Prior for Transfer Learning in GANs

Recent advances in generative adversarial networks (GANs) have shown rem...
research
04/06/2020

Leveraging GANs to Improve Continuous Path Keyboard Input Models

Continuous path keyboard input has higher inherent ambiguity than standa...
research
09/25/2019

Conditional Transferring Features: Scaling GANs to Thousands of Classes with 30

Generative adversarial network (GAN) has greatly improved the quality of...
research
12/27/2022

Knowledge-Guided Data-Centric AI in Healthcare: Progress, Shortcomings, and Future Directions

The success of deep learning is largely due to the availability of large...
research
05/28/2022

Looks Like Magic: Transfer Learning in GANs to Generate New Card Illustrations

In this paper, we propose MAGICSTYLEGAN and MAGICSTYLEGAN-ADA - both inc...

Please sign up or login with your details

Forgot password? Click here to reset