mlsanityData to 14 Oct 2026

Image Augmentations for GAN Training

arXiv 2006.02595 ↗First posted 4 Jun 2020cs.LG cs.CV eess.IVcited by 78

Not on the map. Why?

Abstract (arXiv)

Data augmentations have been widely studied to improve the accuracy and robustness of classifiers. However, the potential of image augmentation in improving GAN models for image synthesis has not been thoroughly investigated in previous studies. In this work, we systematically study the effectiveness of various existing augmentation techniques for GAN training in a variety of settings. We provide insights and guidelines on how to augment images for both vanilla GANs and GANs with regularizations, improving the fidelity of the generated images substantially. Surprisingly, we find that vanilla GANs attain generation quality on par with recent state-of-the-art results if we use augmentations on both real and generated images. When this GAN training is combined with other augmentation-based regularization techniques, such as contrastive loss and consistency regularization, the augmentations further improve the quality of generated images. We provide new state-of-the-art results for conditional generation on CIFAR-10 with both consistency loss and contrastive loss as additional regularizations.

Key facts. Cited by 78 papers in the map (data through 14 Oct 2026). First posted on arXiv 4 Jun 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 34; 2022: 20; 2023: 15; 2024: 4; 2025: 4; 2026: 1.

78 papers.

Sort:

By area:

  1. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,106
  2. 2022Dataset Distillation by Matching Training Trajectoriescited by 442
  3. 2021Dataset Condensation with Differentiable Siamese Augmentationcited by 291
  4. 2021Cross-Modal Contrastive Learning for Text-to-Image Generationcited by 259
  5. 2021Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Modelscited by 252
  6. 2021Projected GANs Converge Fastercited by 208
  7. 2021StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generatorscited by 164
  8. 2021A Good Image Generator Is What You Need for High-Resolution Video Synthesiscited by 163
  9. 2021ViTGAN: Training GANs with Vision Transformerscited by 110
  10. 2021Regularizing Generative Adversarial Networks under Limited Datacited by 107
  11. 2023DREAM: Efficient Dataset Distillation by Representative Matchingcited by 94
  12. 2021Ensembling Off-the-shelf Models for GAN Trainingcited by 86
38 more
  1. 2022Any-resolution Training for High-resolution Image Synthesiscited by 72
  2. 2021Data-Efficient Instance Generation from Instance Discriminationcited by 58
  3. 2022Few Shot Generative Model Adaption via Relaxed Spatial Structural Alignmentcited by 54
  4. 2021Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Datacited by 53
  5. 2022A Closer Look at Few-shot Image Generationcited by 49
  6. 2021Particle Cloud Generation with Message Passing Generative Adversarial Networkscited by 46
  7. 2021Training GANs with Stronger Augmentations via Contrastive Discriminatorcited by 45
  8. 2021One-Shot Generative Domain Adaptationcited by 41
  9. 2021Sketch Your Own GANcited by 37
  10. 2022State-of-the-Art in the Architecture, Methods and Applications of StyleGANcited by 34
  11. 2021Generative Adversarial Registration for Improved Conditional Deformable Templatescited by 31
  12. 2021Dual Contrastive Loss and Attention for GANscited by 30
  13. 2023Zero-shot Generative Model Adaptation via Image-specific Prompt Learningcited by 27
  14. 2022Rewriting Geometric Rules of a GANcited by 25
  15. 2022Towards Diverse and Faithful One-shot Adaption of Generative Adversarial Networkscited by 23
  16. 2022Few-shot Image Generation with Diffusion Modelscited by 21
  17. 2021Data augmentation for deep learning based accelerated MRI reconstruction with limited datacited by 20
  18. 2022A Comprehensive Survey on Data-Efficient GANs in Image Generationcited by 20
  19. 2022Improving GANs with A Dynamic Discriminatorcited by 20
  20. 2022HyperDomainNet: Universal Domain Adaptation for Generative Adversarial Networkscited by 19
  21. 2023A Survey on Generative Modeling with Limited Data, Few Shots, and Zero Shotcited by 19
  22. 2023Domain Re-Modulation for Few-Shot Generative Domain Adaptationcited by 18
  23. 2022FakeCLR: Exploring Contrastive Learning for Solving Latent Discontinuity in Data-Efficient GANscited by 17
  24. 2022FreGAN: Exploiting Frequency Components for Training GANs under Limited Datacited by 16
  25. 2023DomainStudio: Fine-Tuning Diffusion Models for Domain-Driven Image Generation using Limited Datacited by 15
  26. 2022Few-shot Image Generation via Masked Discriminationcited by 12
  27. 2023Phasic Content Fusing Diffusion Model with Directional Distribution Consistency for Few-Shot Model Adaptioncited by 11
  28. 2021Exploring The Effect of High-frequency Components in GANs Trainingcited by 10
  29. 2021Distilling and Transferring Knowledge via cGAN-generated Samples for Image Classification and Regressioncited by 10
  30. 2022StyleDomain: Efficient and Lightweight Parameterizations of StyleGAN for One-shot and Few-shot Domain Adaptationcited by 10
  31. 2021CharacterGAN: Few-Shot Keypoint Character Animation and Reposingcited by 9
  32. 2021ReMix: Towards Image-to-Image Translation with Limited Datacited by 9
  33. 2021Partition-Guided GANscited by 9
  34. 2021Self-Supervised GANs with Label Augmentationcited by 9
  35. 2021Generating Novel Scene Compositions from Single Images and Videoscited by 8
  36. 2024Taming the Tail in Class-Conditional GANs: Knowledge Sharing via Unconditional Training at Lower Resolutionscited by 8
  37. 2022Augmentation-Aware Self-Supervision for Data-Efficient GAN Trainingcited by 7
  38. 2023Unpaired Image-to-Image Translation with Limited Data to Reveal Subtle Phenotypescited by 7

50 most cited of 78 All 78 (JSON)