Research map / Image generation / Super-resolution
Transitional Learning: Exploring the Transition States of Degradation for Blind Super-Resolution
Yuanfei Huang, Jie Li, Yanting Hu, Xinbo Gao, Hua Huang
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 transitional learning with true DoT to primary-degradation baseline (ablation) | baseline SR models · additive degradations N=0 vs N=30 | PSNR gap (dB): 7dB -> 2.6dB | Improved |
| use TLSR instead of IKC | blind SR · Set14 x2 convolutive | PSNR (dB): 32.39 -> 33.08 | Improved |
| use TLSR instead of ZSSR | blind SR · Set14 x2 convolutive | PSNR (dB): 29.45 -> 33.08 | Improved |
| use TLSR instead of IKC for speed | blind SR · Set14 x2 convolutive | inference time (sec): 0.428 -> 0.039 | Improved |
| use TLSR instead of VDN+RCAN | blind SR · Set14 x2 additive | PSNR (dB): 30.69 -> 30.95 | Improved |
| use TLSR instead of DASR | blind SR · Set14 x4 convolutive | PSNR (dB): 28.06 -> 28.34 | Improved |
| use two-stage TLSR_Real instead of VDN+IKC | mixed degradation SR · Urban100 x4 mixed | PSNR (dB): 25.75 -> 26.38 | Improved |
| use TLSR_Real instead of DASR at higher noise | mixed degradation SR · Set14 x4 mixed (blur 0.2, noise 10) | PSNR (dB): 27.07 vs 27.04 | No clear effect |
| use TLSR with single 1-D DoT instead of DASR | anisotropic blur SR · Urban100 x4 anisotropic | PSNR (dB): 24.60 -> 25.84 | Improved |
| use TLSR instead of QGAC | JPEG artifact removal · Classic5 Q=10 | PSNR (dB): 29.84 -> 29.92 | Improved |
| use estimated DoT instead of true DoT (ablation) | TLSR · additive/convolutive SR | PSNR (dB) | No clear effect |
Closest papers by meaning
- End-to-End Alternating Optimization for Real-World Blind Super Resolution (IJCV 2023 · similarity 0.91)
- Finding Discriminative Filters for Specific Degradations in Blind Super-Resolution (NeurIPS 2021 · similarity 0.90)
- Navigating Beyond Dropout: An Intriguing Solution towards Generalizable Image Super Resolution (CVPR 2024 · similarity 0.88)
- Unsupervised Degradation Representation Aware Transform for Real-World Blind Image Super-Resolution (AAAI 2025 · similarity 0.87)
- DORNet: A Degradation Oriented and Regularized Network for Blind Depth Super-Resolution (CVPR 2025 · similarity 0.87)
- CDFormer: When Degradation Prediction Embraces Diffusion Model for Blind Image Super-Resolution (CVPR 2024 · similarity 0.86)
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.
