F-Drop&Match: GANs with a Dead Zone in the High-Frequency Domain
Not on the map. Why?
Abstract (arXiv)
Generative adversarial networks built from deep convolutional neural networks (GANs) lack the ability to exactly replicate the high-frequency components of natural images. To alleviate this issue, we introduce two novel training techniques called frequency dropping (F-Drop) and frequency matching (F-Match). The key idea of F-Drop is to filter out unnecessary high-frequency components from the input images of the discriminators. This simple modification prevents the discriminators from being confused by perturbations of the high-frequency components. In addition, F-Drop makes the GANs focus on fitting in the low-frequency domain, in which there are the dominant components of natural images. F-Match minimizes the difference between real and fake images in the frequency domain for generating more realistic images. F-Match is implemented as a regularization term in the objective functions of the generators; it penalizes the batch mean error in the frequency domain. F-Match helps the generators to fit in the high-frequency domain filtered out by F-Drop to the real image. We experimentally demonstrate that the combination of F-Drop and F-Match improves the generative performance of GANs in both the frequency and spatial domain on multiple image benchmarks.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 24 of its 33 references that are in the map. First posted on arXiv 4 Jun 2021.
Builds on
24 of 33 references.
References · 24
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
- 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
- 2014Deep Learning Face Attributes in the Wildcited by 3,679
- 2016Improved Techniques for Training GANscited by 2,938
- 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
- 2017Improved Training of Wasserstein GANscited by 2,223
- 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
- 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
- 2018Demystifying MMD GANscited by 1,268
- 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
- 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,095
12 more
- 2020Leveraging Frequency Analysis for Deep Fake Image Recognitioncited by 501
- 2019High Frequency Component Helps Explain the Generalization of Convolutional Neural Networkscited by 310
- 2019A Fourier Perspective on Model Robustness in Computer Visioncited by 288
- 2020Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral Distributionscited by 278
- 2018cGANs with Projection Discriminatorcited by 253
- 2017MMD GAN: Towards Deeper Understanding of Moment Matching Networkcited by 251
- 2020Learning in the Frequency Domaincited by 196
- 2019Consistency Regularization for Generative Adversarial Networkscited by 134
- 2020SSD-GAN: Measuring the Realness in the Spatial and Spectral Domainscited by 38
- 2020InfoMax-GAN: Improved Adversarial Image Generation via Information Maximization and Contrastive Learningcited by 37
- 2018On the Structural Sensitivity of Deep Convolutional Networks to the Directions of Fourier Basis Functionscited by 31
- 2020Mimicry: Towards the Reproducibility of GAN Researchcited by 14
Built on by
Citing papers per year: 2023: 1; 2024: 0; 2025: 1.
2 papers.