arXiv (Cornell University) · 2021 · 25 citations · 42 references
ComputingPerforming ArtsComputational ArtComputer VisionReal-time RenderingDeep LearningGenerative ModelingImage SynthesisBenchmark DatasetsFew-shot LearningGenerative Adversarial NetworksGenerative Image SynthesisSelf-supervised AdaptationSelf-supervised Pretraining
Training Generative Adversarial Networks (GAN) on high-fidelity images\nusually requires large-scale GPU-clusters and a vast number of training images.\nIn this paper, we study the few-shot image synthesis task for GAN with minimum\ncomputing cost. We propose a light-weight GAN structure that gains superior\nquality on 1024*1024 resolution. Notably, the model converges from scratch with\njust a few hours of training on a single RTX-2080 GPU, and has a consistent\nperformance, even with less than 100 training samples. Two technique designs\nconstitute our work, a skip-layer channel-wise excitation module and a\nself-supervised discriminator trained as a feature-encoder. With thirteen\ndatasets covering a wide variety of image domains (The datasets and code are\navailable at: https://github.com/odegeasslbc/FastGAN-pytorch), we show our\nmodel's superior performance compared to the state-of-the-art StyleGAN2, when\ndata and computing budget are limited.\n
42
Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren et al. · 2016 · 222.1K citations · Full text
Squeeze-and-Excitation Networks
Jie Hu, Li Shen, Samuel Albanie et al. · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2019 · 12.5K citations
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Richard Zhang, Phillip Isola, Alexei A. Efros et al. · 2018 · 12.4K citations
Momentum Contrast for Unsupervised Visual Representation Learning
Kaiming He, Haoqi Fan, Yuxin Wu et al. · 2020 · 12K citations