Research map / Image generation / Image restoration
Efficient Frequency Domain-Based Transformers for High-Quality Image Deblurring
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 |
|---|---|---|---|
| Use FFTformer instead of NAFNet | Image deblurring · GoPro | PSNR: 33.71 -> 34.21 | Helped |
| Use FFTformer instead of Restormer-local | Image deblurring · HIDE | PSNR: 31.49 -> 31.62 | Helped |
| Use FFTformer instead of DeepRFT+ | Image deblurring · RealBlur-R | PSNR: 39.84 -> 40.11 | Helped |
| Use FFTformer instead of Stripformer | Image deblurring · RealBlur-J | PSNR: 32.48 -> 32.62 | Helped |
| Add FSAS to a model using only the original FFN (ablation) | FFTformer · GoPro ablation | PSNR: 33.19 -> 33.61 | Helped |
| Use DFFN instead of original FFN (ablation) | Model using only the feed-forward component · GoPro ablation | PSNR: 33.19 -> 33.55 | Helped |
| Use DFFN instead of original FFN (ablation) | FSAS-based model · GoPro ablation | PSNR: 33.61 -> 33.73 | Helped |
| Replace FSAS with shifted-window Swin attention (ablation) | Model with DFFN · GoPro ablation | PSNR: 33.46 -> 33.73 | Hurt |
| Use FSAS only in the decoder rather than in both encoder and decoder (ablation) | FFTformer · GoPro asymmetric architecture | PSNR: 33.56 -> 33.73 | Helped |
Closest papers by meaning
- Beyond Spatial Domain: Cross-domain Promoted Fourier Convolution Helps Single Image Dehazing (AAAI 2025 · similarity 0.88)
- Efficient RAW Image Deblurring with Adaptive Frequency Modulation (NeurIPS 2025 · similarity 0.88)
- Efficient Visual State Space Model for Image Deblurring (CVPR 2025 · similarity 0.87)
- DeblurDiff: Real-Word Image Deblurring with Generative Diffusion Models (NeurIPS 2025 · similarity 0.87)
- Boosting ViT-based MRI Reconstruction from the Perspectives of Frequency Modulation, Spatial Purification, and Scale Diversification (AAAI 2025 · similarity 0.86)
- Misalignment-Robust Frequency Distribution Loss for Image Transformation (CVPR 2024 · similarity 0.86)
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.