Research map / Image generation / Super-resolution
Blind Image Super-Resolution: A Survey and Beyond
Anran Liu, Yihao Liu, Jinjin Gu, Yu Qiao, Chao Dong
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 NPBSR kernel estimation with KernelGAN, keeping ZSSR | ZSSR · DIV2KRK (x2) | PSNR (dB): 29.37 -> 30.36 | Improved |
| use ground-truth kernel instead of KernelGAN estimate | ZSSR · DIV2KRK (x2) | PSNR (dB): 30.36 -> 32.44 | Improved |
| use ground-truth kernel instead of KernelGAN estimate | SRMD · DIV2KRK (x4) | PSNR (dB): 25.71 -> 27.38 | Improved |
| pair NPBSR kernel estimate with SRMD instead of bicubic | SRMD · DIV2KRK (x2) | PSNR (dB): 28.73 -> 25.51 | Hurt |
| joint kernel estimation and SR (DAN) instead of ZSSR with KernelGAN | DAN · DIV2KRK (x2) | PSNR (dB): 30.36 -> 32.56 | Improved |
| iterative corrector trained jointly end-to-end (DAN) instead of separate IKC | DAN · Set14 (x2) | PSNR (dB): 32.82 -> 33.07 | Improved |
| joint training of corrector and SR network (DAN) instead of IKC | DAN · BSD100 (x3) | PSNR (dB): 28.56 -> 28.94 | Improved |
| joint training (DAN) instead of IKC | DAN · Set14 (x4) | PSNR (dB): 28.26 -> 28.43 | Improved |
| domain-gap aware training (DASR) instead of FSSR | DASR · RealSR | PSNR (dB): 25.99 -> 26.78 | Improved |
| domain-gap aware training (DASR) instead of FSSR | DASR · RealSR | LPIPS: 0.27 -> 0.23 | Improved |
| domain-gap aware training (DASR) instead of CinCGAN | DASR · AIM (real-world SR) | LPIPS: 0.46 -> 0.34 | Improved |
| domain-gap aware training (DASR) instead of FSSR | DASR · AIM (real-world SR) | PSNR (dB): 20.82 -> 21.60 | Mixed |
Closest papers by meaning
- End-to-End Alternating Optimization for Real-World Blind Super Resolution (IJCV 2023 · similarity 0.92)
- Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (ICCV 2021 · similarity 0.91)
- Deep Learning for Image Super-Resolution: A Survey (TPAMI 2020 · similarity 0.91)
- Deep Constrained Least Squares for Blind Image Super-Resolution (CVPR 2022 · similarity 0.91)
- Navigating Beyond Dropout: An Intriguing Solution towards Generalizable Image Super Resolution (CVPR 2024 · similarity 0.90)
- Unsupervised Degradation Representation Aware Transform for Real-World Blind Image Super-Resolution (AAAI 2025 · 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.
