English

Few-shot Image Generation via Cross-domain Correspondence

Computer Vision and Pattern Recognition 2021-04-15 v1 Graphics Machine Learning

Abstract

Training generative models, such as GANs, on a target domain containing limited examples (e.g., 10) can easily result in overfitting. In this work, we seek to utilize a large source domain for pretraining and transfer the diversity information from source to target. We propose to preserve the relative similarities and differences between instances in the source via a novel cross-domain distance consistency loss. To further reduce overfitting, we present an anchor-based strategy to encourage different levels of realism over different regions in the latent space. With extensive results in both photorealistic and non-photorealistic domains, we demonstrate qualitatively and quantitatively that our few-shot model automatically discovers correspondences between source and target domains and generates more diverse and realistic images than previous methods.

Keywords

Cite

@article{arxiv.2104.06820,
  title  = {Few-shot Image Generation via Cross-domain Correspondence},
  author = {Utkarsh Ojha and Yijun Li and Jingwan Lu and Alexei A. Efros and Yong Jae Lee and Eli Shechtman and Richard Zhang},
  journal= {arXiv preprint arXiv:2104.06820},
  year   = {2021}
}

Comments

CVPR 2021

R2 v1 2026-06-24T01:09:36.554Z