mlsanityData to 7 Oct 2026

ComGAN: Toward GANs Exploiting Multiple Samples

arXiv 2304.12098 ↗First posted 24 Apr 2023cs.LG cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

In this paper, we propose ComGAN(ComparativeGAN) which allows the generator in GANs to refer to the semantics of comparative samples(e.g. real data) by comparison. ComGAN generalizes relativistic GANs by using arbitrary architecture and mostly outperforms relativistic GANs in simple input-concatenation architecture. To train the discriminator in ComGAN, we also propose equality regularization, which fits the discriminator to a neutral label for equally real or fake samples. Equality regularization highly boosts the performance of ComGAN including WGAN while being exceptionally simple compared to existing regularizations. Finally, we generalize comparative samples fixed to real data in relativistic GANs toward fake data and show that such objectives are sound in both theory and practice. Our experiments demonstrate superior performances of ComGAN and equality regularization, achieving the best FIDs in 7 out of 8 cases of different losses and data against ordinary GANs and relativistic GANs.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 37 of its 46 references that are in the map. First posted on arXiv 24 Apr 2023.

Builds on

37 of 46 references.

References · 37

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  3. 2022Generative Adversarial Networkscited by 5,603
  4. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  5. 2016Improved Techniques for Training GANscited by 2,938
  6. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  7. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  8. 2017Improved Training of Wasserstein GANscited by 2,223
  9. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
  10. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  11. 2018ESRGAN: Enhanced Super-Resolution Generative Adversarial Networkscited by 1,250
  12. 2017Mixed Precision Trainingcited by 1,135
25 more
  1. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,095
  2. 2016Least Squares Generative Adversarial Networkscited by 1,000
  3. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 925
  4. 2018Which Training Methods for GANs do actually Converge?cited by 672
  5. 2017Unsupervised Image-to-Image Translation Networkscited by 602
  6. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 504
  7. 2017Towards Principled Methods for Training Generative Adversarial Networkscited by 460
  8. 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 425
  9. 2019DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Bettercited by 361
  10. 2017Are GANs Created Equal? A Large-Scale Studycited by 267
  11. 2020Deep Unfolding Network for Image Super-Resolutioncited by 253
  12. 2018The relativistic discriminator: a key element missing from standard GANcited by 234
  13. 2019Consistency Regularization for Generative Adversarial Networkscited by 134
  14. 2018Improving GANs Using Optimal Transportcited by 122
  15. 2019Image Super-Resolution by Neural Texture Transfercited by 119
  16. 2017The Cramer Distance as a Solution to Biased Wasserstein Gradientscited by 116
  17. 2021Regularizing Generative Adversarial Networks under Limited Datacited by 105
  18. 2017PacGAN: The power of two samples in generative adversarial networkscited by 104
  19. 2018CrossNet: An End-to-end Reference-based Super Resolution Network using Cross-scale Warpingcited by 90
  20. 2020Improved Consistency Regularization for GANscited by 67
  21. 2022StudioGAN: A Taxonomy and Benchmark of GANs for Image Synthesiscited by 38
  22. 2021Dual Contrastive Loss and Attention for GANscited by 30
  23. 2017McGan: Mean and Covariance Feature Matching GANcited by 21
  24. 2020Self-Adversarial Learning with Comparative Discrimination for Text Generationcited by 11
  25. 2020Regularization Methods for Generative Adversarial Networks: An Overview of Recent Studiescited by 5

Built on by

Not cited yet.