D3T-GAN: Data-Dependent Domain Transfer GANs for Few-shot Image Generation

TL;DR

D3T-GAN introduces a novel self-supervised transfer approach for few-shot image generation, effectively transferring knowledge between source and target GANs to improve image quality with state-of-the-art results.

Contribution

The paper proposes D3T-GAN, a new transfer scheme that independently transfers knowledge between generators and discriminators in few-shot GANs.

Findings

Achieves state-of-the-art FID scores on benchmark datasets.

Improves the realism and diversity of generated images.

Effective knowledge transfer enhances few-shot image generation quality.

Abstract

As an important and challenging problem, few-shot image generation aims at generating realistic images through training a GAN model given few samples. A typical solution for few-shot generation is to transfer a well-trained GAN model from a data-rich source domain to the data-deficient target domain. In this paper, we propose a novel self-supervised transfer scheme termed D3T-GAN, addressing the cross-domain GANs transfer in few-shot image generation. Specifically, we design two individual strategies to transfer knowledge between generators and discriminators, respectively. To transfer knowledge between generators, we conduct a data-dependent transformation, which projects and reconstructs the target samples into the source generator space. Then, we perform knowledge transfer from transformed samples to generated samples. To transfer knowledge between discriminators, we design a multi-level discriminant knowledge distillation from the source discriminator to the target discriminator on both the real and fake samples. Extensive experiments show that our method improve the quality of generated images and achieves the state-of-the-art FID scores on commonly used datasets.