EUVIMEDEUVIMED entdecken
Crossref

WeditGAN: Few-Shot Image Generation via Latent Space Relocation

Yuxuan Duan, Li Niu, Yan Hong, Liqing Zhang

Proceedings of the AAAI Conference on Artificial Intelligence · 24.03.2024 · 38 (2) · 1653-1661

DOI: 10.1609/aaai.v38i2.27932 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 24.03.2024. Verfasst von Yuxuan Duan und 3 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In few-shot image generation, directly training GAN models on just a handful of images faces the risk of overfitting. A popular solution is to transfer the models pretrained on large source domains to small target ones. In this work, we introduce WeditGAN, which realizes model transfer by editing the intermediate latent codes w in StyleGANs with learned constant offsets (delta w), discovering and constructing target latent spaces via simply relocating the distribution of source latent spaces. The established one-to-one mapping between latent spaces can naturally prevents mode collapse and overfitting. Besides, we also propose variants of WeditGAN to further enhance the relocation process by regularizing the direction or finetuning the intensity of delta w. Experiments on a collection of widely used source/target datasets manifest the capability of WeditGAN in generating realistic and diverse images, which is simple yet highly effective in the research area of few-shot image generation. Codes are available at https://github.com/Ldhlwh/WeditGAN.

Zitieren

APA

Duan, Y., Niu, L., Hong, Y., & Zhang, L. (2024). WeditGAN: Few-Shot Image Generation via Latent Space Relocation. Proceedings of the AAAI Conference on Artificial Intelligence, 38(2), 1653-1661. https://doi.org/10.1609/aaai.v38i2.27932

Vancouver

Duan Y, Niu L, Hong Y, Zhang L. WeditGAN: Few-Shot Image Generation via Latent Space Relocation. Proceedings of the AAAI Conference on Artificial Intelligence. 2024;38(2):1653-1661. doi:10.1609/aaai.v38i2.27932

BibTeX
@article{Duan2024,
  author = {Duan, Yuxuan and Niu, Li and Hong, Yan and Zhang, Liqing},
  title = {WeditGAN: Few-Shot Image Generation via Latent Space Relocation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2024},
  volume = {38},
  number = {2},
  pages = {1653--1661},
  doi = {10.1609/aaai.v38i2.27932}
}
Kostenlos starten →