Research map / Image generation / Image restoration
Dual-Domain Attention for 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 |
|---|---|---|---|
| Add SAM to the baseline (ablation) | Small baseline network · GoPro | PSNR: 31.42 -> 31.82 | Helped |
| Add FAM to the baseline (ablation) | Small baseline network · GoPro | PSNR: 31.42 -> 32.17 | Helped |
| Add both SAM and FAM (ablation) | Small baseline network · GoPro | PSNR: 31.42 -> 32.29 | Helped |
| Use Tanh rather than Softmax for SAM weights (ablation) | SAM · GoPro | PSNR: 31.66 -> 31.77 | Helped |
| Increase SAM group count from 1 to 16 (ablation) | SAM · GoPro | PSNR: 31.62 -> 31.80 | Helped |
| Replace context-aware SAM weights with standard group convolution (ablation) | SAM · GoPro | PSNR: 31.80 -> 31.52 | Hurt |
| Add the convolution that mixes SAM groups (ablation) | SAM · GoPro | PSNR: 31.80 -> 31.82 | Helped |
| Replace SAM with global self-attention (ablation) | Small baseline network · GoPro | PSNR: 31.55 -> 31.50 | Hurt |
| Replace SAM with MDTA (ablation) | Small baseline network · GoPro | PSNR: 31.55 -> 31.34 | Hurt |
| Increase the number of local FAM filters (ablation) | FAM · GoPro | PSNR: 31.48 -> 31.69 | Helped |
| Use dual-domain attention rather than the baseline | DDANet · GoPro test | PSNR: 32.92 -> 33.07 | Helped |
| Use DDANet instead of Restormer | GoPro test inference · GoPro test | inference time (s): 1.218 -> 0.247 | Helped |
Closest papers by meaning
- Efficient Frequency Domain-Based Transformers for High-Quality Image Deblurring (CVPR 2023 · similarity 0.90)
- Beyond Spatial Domain: Cross-domain Promoted Fourier Convolution Helps Single Image Dehazing (AAAI 2025 · similarity 0.89)
- Wavelet-Assisted Multi-Frequency Attention Network for Pansharpening (AAAI 2025 · similarity 0.85)
- Efficient RAW Image Deblurring with Adaptive Frequency Modulation (NeurIPS 2025 · similarity 0.85)
- Learning 1D Causal Visual Representation with De-focus Attention Networks (NeurIPS 2024 · similarity 0.84)
- Prior-guided Hierarchical Harmonization Network for Efficient Image Dehazing (AAAI 2025 · 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.