Research map / Image generation / GANs
SSD-GAN: Measuring the Realness in the Spatial and Spectral Domains
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add a spectral classifier to the discriminator | StyleGAN · FFHQ | FID: 4.40 -> 4.06 | Helped |
| Add a spectral classifier to the discriminator | StyleGAN · FFHQ | PPL full: 234.0 -> 229.1 | Helped |
| Add a spectral classifier to the discriminator | StyleGAN · FFHQ | PPL end: 195.9 -> 189.7 | Helped |
| Add a spectral classifier to the discriminator | SNGAN · CIFAR100-32 | FID: 22.61 -> 19.28 | Helped |
| Add a spectral classifier to the discriminator | SNGAN · STL10-48 | FID: 39.56 -> 36.41 | Helped |
| Add a spectral classifier to the discriminator | SNGAN · LSUN-128 | FID: 25.87 -> 15.17 | Helped |
| Add spectral regularization to the generator loss | SNGAN · CIFAR100-32 | FID: 22.61 -> 21.39 | Helped |
| Add spectral regularization to the generator loss | SNGAN · STL10-48 | FID: 39.56 -> 38.16 | Helped |
| Add spectral regularization to the generator loss | SNGAN · LSUN-128 | FID: 25.87 -> 16.95 | Helped |
| Replace average-pooling with DWT for downsampling | SNGAN · CIFAR100-32 | FID: 22.61 -> 25.95 | Hurt |
| Replace average-pooling with DWT for downsampling | SNGAN · STL10-48 | FID: 39.56 -> 40.46 | Hurt |
| Replace average-pooling with DWT for downsampling | SNGAN · LSUN-128 | FID: 25.87 -> 148.05 | Hurt |
Closest papers by meaning
- On the Frequency Bias of Generative Models (NeurIPS 2021 · similarity 0.89)
- Fourier Spectrum Discrepancies in Deep Network Generated Images (NeurIPS 2020 · similarity 0.88)
- Spatial Frequency Bias in Convolutional Generative Adversarial Networks (AAAI 2022 · similarity 0.87)
- Spectral Distribution Aware Image Generation (AAAI 2021 · similarity 0.87)
- SeD: Semantic-Aware Discriminator for Image Super-Resolution (CVPR 2024 · similarity 0.85)
- The GAN is dead; long live the GAN! A Modern GAN Baseline (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.