Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
This addresses the challenge of high computational cost and data scarcity in GAN training for image synthesis, offering a more accessible solution for researchers and practitioners with limited resources.
The paper tackles the problem of training Generative Adversarial Networks (GANs) for high-fidelity image synthesis with limited data and computing resources, achieving superior quality at 1024*1024 resolution by converging from scratch in a few hours on a single GPU with less than 100 training samples.
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*1024 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.