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

Abstract

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.

ICLR-2021

References in corpus (8)

• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Energy-based Generative Adversarial Network
• Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs
• Precise Recovery of Latent Vectors from Generative Adversarial Networks
• Self-supervised GAN: Analysis and Improvement with Multi-class Minimax Game
• Small-GAN: Speeding Up GAN Training Using Core-sets
• Improving the Speed and Quality of GAN by Adversarial Training
• DepthwiseGANs: Fast Training Generative Adversarial Networks for Realistic Image Synthesis

Cited by in corpus (8)

• SinGAN-Seg: Synthetic training data generation for medical image segmentation
• Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models
• medigan: a Python library of pretrained generative models for medical image synthesis
• Generative Artificial Intelligence Meets Synthetic Aperture Radar: A Survey
• Neural stochastic differential equations for particle dispersion in large-eddy simulations of homogeneous isotropic turbulence
• GANravel: User-Driven Direction Disentanglement in Generative Adversarial Networks
• Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes
• DEff-GAN: Diverse Attribute Transfer for Few-Shot Image Synthesis