Research map / Image generation / GANs
Spectral Distribution Aware Image Generation
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 the spectral discriminator | DCGAN · FFHQ 64² | spectral difference (SD): 1.293 -> 0.042 | Helped |
| Add the spectral discriminator | DCGAN · FFHQ 64² | FID: 15.257 -> 15.591 | Hurt |
| Add the spectral discriminator | LSGAN · FFHQ 64² | cloaking score (CS): 0.04 -> 0.86 | Helped |
| Add the spectral discriminator | LSGAN · FFHQ 128² | spectral difference (SD): 3.786 -> 0.074 | Helped |
| Add the spectral discriminator | LSGAN · FFHQ 256² | cloaking score (CS): 0.00 -> 0.46 | Helped |
| Add the spectral discriminator | StyleGAN2 · FFHQ 1024² | FID: 2.733 -> 3.326 | Hurt |
| Add the spectral discriminator during finetuning | StyleGAN2 · FFHQ 1024² | spectral difference (SD): 32.250 -> 5.978 | Helped |
| Use modified upsampling and add high-resolution convolutional layers (ablation) | DCGAN · FFHQ 64² | FID: 17.749 -> 15.257 | Helped |
| Use modified upsampling and add high-resolution convolutional layers (ablation) | DCGAN · FFHQ 64² | cloaking score (CS): 0.35 -> 0.16 | Hurt |
| Use modified upsampling and add high-resolution convolutional layers (ablation) | LSGAN · FFHQ 64² | FID: 18.423 -> 15.518 | Helped |
| Use modified upsampling and add high-resolution convolutional layers (ablation) | LSGAN · FFHQ 64² | cloaking score (CS): 0.28 -> 0.04 | Hurt |
| Use spectral regularization toward the average real spectrum | DCGAN · FFHQ 64² | FID: 15.257 -> 29.875 | Hurt |
Closest papers by meaning
- Spectrum Translation for Refinement of Image Generation (STIG) Based on Contrastive Learning and Spectral Filter Profile (AAAI 2024 · similarity 0.88)
- On the Frequency Bias of Generative Models (NeurIPS 2021 · similarity 0.88)
- A Closer Look at Fourier Spectrum Discrepancies for CNN-Generated Images Detection (CVPR 2021 · similarity 0.88)
- Any-Resolution AI-Generated Image Detection by Spectral Learning (CVPR 2025 · similarity 0.87)
- Secret Lies in Color: Enhancing AI-Generated Images Detection with Color Distribution Analysis (CVPR 2025 · similarity 0.86)
- Enhancing Consistency-Based Image Generation via Adversarialy-Trained Classification and Energy-Based Discrimination (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.