Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Data

2021/11/12 by Liming Jiang, Jiang, Liming, Bo Dai +5 · 6 citations
Computer Science · #Advanced Image Processing Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Image and Signal Denoising Methods #Machine Learning (cs.LG) #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.2111.06849

NeurIPS 2021. Code: https://github.com/EndlessSora/DeceiveD Project page: https://www.mmlab-ntu.com/project/apa/index.html

arxiv created 2021/11/12 · openalex publication_date 2021/11/12 · arxiv updated 2021/11/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Generative adversarial networks (GANs) typically require ample data for training in order to synthesize high-fidelity images. Recent studies have shown that training GANs with limited data remains formidable due to discriminator overfitting, the underlying cause that impedes the generator's convergence. This paper introduces a novel strategy called Adaptive Pseudo Augmentation (APA) to encourage healthy competition between the generator and the discriminator. As an alternative method to existing approaches that rely on standard data augmentations or model regularization, APA alleviates overfitting by employing the generator itself to augment the real data distribution with generated images, which deceives the discriminator adaptively. Extensive experiments demonstrate the effectiveness of APA in improving synthesis quality in the low-data regime. We provide a theoretical analysis to examine the convergence and rationality of our new training strategy. APA is simple and effective. It can be added seamlessly to powerful contemporary GANs, such as StyleGAN2, with negligible computational cost.

Citations

• Divergence measures based on the Shannon entropy
• Amortised MAP Inference for Image Super-resolution
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis
• Image Augmentations for GAN Training
• Learning Multiple Layers of Features from Tiny Images

Cited by

• UCD: Unconditional Discriminator Promotes Nash Equilibrium in GANs
• Intra-Cluster Mixup: An Effective Data Augmentation Technique for Complementary-Label Learning
• Styleclone: Face Stylization with Diffusion Based Data Augmentation
• Score Augmentation for Diffusion Models
• Hyperbolic Dual Feature Augmentation for Open-Environment
• Mosaic: Data-Free Knowledge Distillation via Mixture-of-Experts for Heterogeneous Distributed Environments

Related

• Training Generative Adversarial Networks with Limited Data
• Differentiable Augmentation for Data-Efficient GAN Training
• Improving GANs with A Dynamic Discriminator
• On Data Augmentation for GAN Training
• Training GANs with Stronger Augmentations via Contrastive Discriminator