Research map / Image generation / Super-resolution
Wavelet Domain Style Transfer for an Effective Perception-Distortion Tradeoff in Single Image Super-Resolution
Xin Deng, Ren Yang, Mai Xu, Pier Luigi Dragotti
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 |
|---|---|---|---|
| apply WDST to the LH sub-band (ablation) | WDST · BSD100 | NRQM: 7.8490 -> 8.5948 | Improved |
| apply WDST to the HL sub-band (ablation) | WDST · BSD100 | NRQM: 7.8343 -> 8.5948 | Improved |
| apply WDST to the HH sub-band (ablation) | WDST · BSD100 | NRQM: 8.0542 -> 8.5948 | Improved |
| apply WDST to the HH sub-band (ablation) | WDST · BSD100 | PSNR: 26.96 -> 26.82 | Hurt |
| apply WDST to the LH sub-band (ablation) | WDST · BSD100 | PSNR: 27.19 -> 26.82 | Hurt |
| swap wavelet filter between haar, db2, bior2.2, bior4.4 etc. (ablation) | WDST · Set14 | NRQM: 7.5109 to 7.7442 | No clear effect |
| replace low-frequency sub-band of A_p by that of A_o (ablation) | A_p · BSD100 | PSNR: 24.58 -> 25.68 | Improved |
| replace low-frequency sub-band of A_p by that of A_o (ablation) | A_p · BSD100 | NRQM: 8.8007 -> 8.7775 | No clear effect |
| replace low-frequency sub-band of A_o by that of A_p (ablation) | A_o · BSD100 | PSNR: 27.80 -> 26.57 | Hurt |
| perform style transfer in wavelet domain with sub-band separation instead of pixel domain (Deng) | WDST · Set5 | PSNR: 31.14 -> 31.46 | Improved |
| perform style transfer in wavelet domain instead of pixel domain (Deng) | WDST · Set5 | NRQM: 7.0022 -> 7.5180 | Improved |
| use worst-case inputs SRResNet-MSE and SRGAN-vgg54 instead of EDSR and CX | WDST · BSD100 | PSNR/NRQM: 26.56 dB/8.5005 vs Deng 26.46 dB/8.4452 | No clear effect |
Closest papers by meaning
- Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts (CVPR 2024 · similarity 0.90)
- SRFeat: Single Image Super-Resolution with Feature Discrimination (ECCV 2018 · similarity 0.88)
- Auto-Encoded Supervision for Perceptual Image Super-Resolution (CVPR 2025 · similarity 0.85)
- GSDD: Generative Space Dataset Distillation for Image Super-resolution (AAAI 2024 · similarity 0.85)
- Exploring Semantic Feature Discrimination for Perceptual Image Super-Resolution and Opinion-Unaware No-Reference Image Quality Assessment (CVPR 2025 · similarity 0.85)
- SeD: Semantic-Aware Discriminator for Image Super-Resolution (CVPR 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.
