Research map / Image generation / Image restoration
FouriDown: Factoring Down-Sampling into Shuffling and Superposing
Qi Zhu, Man Zhou, Jie Huang, Naishan Zheng, Hongzhi Gao, Chongyi Li, Yuan Xu, Feng Zhao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| replace max-pooling down-sampling with FouriDown | SID · LOL | PSNR: 21.46 -> 23.28 | Improved |
| replace strided down-sampling with FouriDown | DRBN · LOL | PSNR: 19.92 -> 21.64 | Improved |
| replace down-sampling with FouriDown | DRBN · Huawei | PSNR: 20.21 -> 21.46 | Improved |
| add ideal low-pass filter before down-sampling (ablation) | DRBN · LOL | PSNR: 19.92 -> 18.91 | Hurt |
| add Gaussian filter before down-sampling | SID · LOL | PSNR: 21.46 -> 21.78 | Improved |
| add ideal low-pass filter before down-sampling (ablation) | SID · LOL | PSNR: 21.46 -> 20.74 | Hurt |
| replace down-sampling with FouriDown | MPRNet · HIDE | PSNR: 30.04 -> 30.25 | Improved |
| replace down-sampling with FouriDown | DeepDeblur · DVD | PSNR: 29.32 -> 29.44 | Improved |
| replace down-sampling with FouriDown | Restormer · SIDD | PSNR: 39.41 -> 39.47 | No clear effect |
| replace down-sampling with FouriDown | MIRNet · DND | PSNR: 39.41 -> 39.56 | Improved |
| replace down-sampling with FouriDown | MSBDN · RESIDE Outdoor | PSNR: 28.88 -> 29.21 | Improved |
| replace down-sampling with FouriDown | GridNet · RESIDE Indoor | PSNR: 30.16 -> 30.42 | Improved |
Closest papers by meaning
- DiffFNO: Diffusion Fourier Neural Operator (CVPR 2025 · similarity 0.84)
- Improving the Learning Capability of Small-size Image Restoration Network by Deep Fourier Shifting (NeurIPS 2024 · similarity 0.84)
- IV-mixed Sampler: Leveraging Image Diffusion Models for Enhanced Video Synthesis (ICLR 2025 · similarity 0.84)
- EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models (ICLR 2024 · similarity 0.84)
- Multi-Cross Sampling and Frequency-Division Reconstruction for Image Compressed Sensing (AAAI 2024 · similarity 0.84)
- Dual-Rate Diffusion: Accelerating diffusion models with an interleaved heavy-light network (NeurIPS 2026 · similarity 0.84)
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.
