Research map / Image generation / Super-resolution
HDW-SR: High-Frequency Guided Diffusion Model based on Wavelet Decomposition for Image Super-Resolution
Chao Yang, Boqian Zhang, Jinghao Xu, Guang Jiang
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 DWT with CNN down/upsampling (ablation) | HDW-Net · RealSR (x4) | PSNR: 25.16 -> 25.71 | Improved |
| add high-frequency guidance from PreSR via HE-Net (vs no guidance, self-attention) (ablation) | HDW-Net · RealSR (x4) | PSNR: 22.15 -> 25.71 | Improved |
| take guidance from HE-Net rather than HA-Net (ablation) | HDW-Net · RealSR (x4) | PSNR: 24.39 -> 25.71 | Improved |
| use Dynamic Thresholding Block instead of Top-k (ablation) | DFA encoder · RealSR (x4) | PSNR: 25.37 -> 25.71 | Improved |
| use DTB instead of Top-k (ablation) | DFA encoder · RealSR (x4) | FLOPs (G): 172 -> 134 | Improved |
| use DTB instead of Top-k (ablation) | DFA encoder · RealSR (x4) | E2E time (s/image): 1.765 -> 1.435 | Improved |
| set loss weight beta to 0.2 instead of 0.1 (ablation) | HDW-SR · DIV2K | PSNR: 18.39 -> 24.52 | Improved |
| set beta to 0.2 instead of 0.5 (ablation) | HDW-SR · DIV2K | PSNR: 10.30 -> 24.52 | Improved |
| HDW-SR versus DiT-SR | HDW-SR · RealSR (x4) | PSNR: 25.31 -> 25.71 | Improved |
| HDW-SR versus DiT-SR on NIQE | HDW-SR · RealSR (x4) | NIQE: 5.36 vs 5.39 | No clear effect |
| HDW-SR versus ResShift on PSNR | HDW-SR · DIV2K | PSNR: 24.69 vs 24.52 | Hurt |
| HDW-SR versus GAN methods on no-reference metrics | HDW-SR · DIV2K | MUSIQ: 61.20 -> 69.68 | Improved |
Closest papers by meaning
- ResDiff: Combining CNN and Diffusion Model for Image Super-resolution (AAAI 2024 · similarity 0.89)
- Accelerated Diffusion via High-Low Frequency Decomposition for Pan-Sharpening (AAAI 2025 · similarity 0.89)
- SinSR: Diffusion-Based Image Super-Resolution in a Single Step (CVPR 2024 · similarity 0.88)
- Arbitrary-steps Image Super-resolution via Diffusion Inversion (CVPR 2025 · similarity 0.88)
- TSD-SR: One-Step Diffusion with Target Score Distillation for Real-World Image Super-Resolution (CVPR 2025 · similarity 0.88)
- Effective Diffusion Transformer Architecture for Image Super-Resolution (AAAI 2025 · similarity 0.88)
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.
