Domain Adaptation for Learning Generator from Paired Few-Shot Data

TL;DR

This paper introduces PFS-GAN, a novel generative model that leverages few-shot learning and domain transfer to produce diverse, high-quality target domain images with limited data, by disentangling content and appearance features.

Contribution

The paper presents a new PFS-GAN model that effectively transfers knowledge across domains using paired few-shot data, improving generator quality and diversity.

Findings

Outperforms baseline methods in quality and diversity of generated images.

Successfully disentangles content and appearance features across domains.

Achieves higher quantitative scores on target domain generation tasks.

Abstract

We propose a Paired Few-shot GAN (PFS-GAN) model for learning generators with sufficient source data and a few target data. While generative model learning typically needs large-scale training data, our PFS-GAN not only uses the concept of few-shot learning but also domain shift to transfer the knowledge across domains, which alleviates the issue of obtaining low-quality generator when only trained with target domain data. The cross-domain datasets are assumed to have two properties: (1) each target-domain sample has its source-domain correspondence and (2) two domains share similar content information but different appearance. Our PFS-GAN aims to learn the disentangled representation from images, which composed of domain-invariant content features and domain-specific appearance features. Furthermore, a relation loss is introduced on the content features while shifting the appearance features to increase the structural diversity. Extensive experiments show that our method has better quantitative and qualitative results on the generated target-domain data with higher diversity in comparison to several baselines.