Research map / Image generation / Super-resolution
Uncertainty Learning in Kernel Estimation for Multi-stage Blind Image Super-Resolution
Zhenxuan Fang, Weisheng Dong, Xin Li, Jinjian Wu, Leida Li, Guangming Shi
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 |
|---|---|---|---|
| add uncertainty learning to kernel estimation (ablation) | kernel estimation network · average over 9 kernels, noise 20 | kernel MAE: 0.252 -> 0.172 | Improved |
| add uncertainty learning to kernel estimation (ablation) | kernel estimation network · average over 9 kernels, noise 0 | kernel MAE: 0.190 -> 0.146 | Improved |
| add uncertainty learning to kernel estimation (ablation) | KULNet · Set14 x4 (noise 0) | PSNR: 27.49 -> 27.81 | Improved |
| add uncertainty learning to kernel estimation (ablation) | KULNet · Urban100 x2 (noise 0) | PSNR: 27.75 -> 28.16 | Improved |
| add uncertainty learning to kernel estimation (ablation) | KULNet · Set5 x2 (noise 20) | PSNR: 29.27 -> 29.54 | Improved |
| add KL regularization (lambda 0 -> 0.001) (ablation) | KULNet · Set14 x4 | PSNR: 26.49 -> 26.76 | Improved |
| add KL regularization (lambda 0 -> 0.001) (ablation) | KULNet · BSD100 x4 | PSNR: 25.95 -> 26.12 | Improved |
| raise KL weight too high (0.001 -> 1) (ablation) | KULNet · Set14 x4 | PSNR: 26.76 -> 25.89 | Hurt |
| raise KL weight from 0.001 to 0.1 (ablation) | KULNet · BSD100 x4 | PSNR: 26.12 -> 25.62 | Hurt |
| add dense connections (ablation) | KULNet · Set14 x4 | PSNR: 0.13 dB | Improved |
| increase number of stages T (ablation) | KULNet · Set14 x2 and x4 | PSNR | Improved |
| replace MANet with KULNet | blind SR · Urban100 x4 (noise 0) | PSNR: 24.49 -> 25.07 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- KOALAnet — KOALAnet: Blind Super-Resolution Using Kernel-Oriented Adaptive Local Adjustment (CVPR 2021) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Deep Constrained Least Squares for Blind Image Super-Resolution (CVPR 2022 · similarity 0.91)
- Unfolding the Alternating Optimization for Blind Super Resolution (NeurIPS 2020 · similarity 0.91)
- Spectrum-to-Kernel Translation for Accurate Blind Image Super-Resolution (NeurIPS 2021 · similarity 0.91)
- Learning the Non-Differentiable Optimization for Blind Super-Resolution (CVPR 2021 · similarity 0.91)
- Blind Super-Resolution With Iterative Kernel Correction (CVPR 2019 · similarity 0.91)
- A Dynamic Kernel Prior Model for Unsupervised Blind Image Super-Resolution (CVPR 2024 · similarity 0.90)
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.
