Training Generative Adversarial Networks (GAN) on high-fidelity images usually requires large-scale GPU-clusters and a vast number of training images. In this paper, we study the few-shot image synthesis task for GAN with minimum computing cost. We propose a light-weight GAN structure that gains superior quality on 1024^2 resolution. Notably, the model converges from scratch with just a few hours of training on a single RTX-2080 GPU, and has a consistent performance, even with less than 100 training samples. Two technique designs constitute our work, a skip-layer channel-wise excitation module and a self-supervised discriminator trained as a feature-encoder. With thirteen datasets covering a wide variety of image domains (The datasets and code are available at https://github.com/odegeasslbc/FastGAN-pytorch), we show our model's superior performance compared to the state-of-the-art StyleGAN2, when data and computing budget are limited.

A computational-efficient GAN for few-shot hi-fi image dataset (converge on single gpu with few hours' training, on 1024 resolution sub-hundred images).

Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis

Abstract of paper. Add TeX formulas using the following formats: $In-line Formula$ or $$Block Formula$$

Search author profile by first, middle and last name or email address. If the profile is not found, you can add the author by completing first, middle, and last names as well as author email address.

See the ICLR Code of Ethics: https://iclr.cc/public/CodeOfEthics

Upload a PDF file that ends with .pdf. The maximum file size is 50MB. Note: optional before the abstract submission deadline.

All supplementary material must be self-contained and zipped into a single file. Note that supplementary material will be visible to reviewers and the public throughout and after the review period, and ensure all material is anonymized. The maximum file size is 100MB.

Title of paper. Add TeX formulas using the following formats: $In-line Formula$ or $$Block Formula$$

Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis

Bingchen Liu, Yizhe Zhu, Kunpeng Song, Ahmed Elgammal