Research map / Image generation / Super-resolution
Frequency Consistent Adaptation for Real World Super Resolution
Xiaozhong Ji, Guangpin Tao, Yun Cao, Ying Tai, Tong Lu, Chengjie Wang, Jilin Li, Feiyue 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 |
|---|---|---|---|
| adapt RCAN training pairs to the estimated real degradation with FCA | RCAN · ISO.1 | PSNR: 23.56 -> 27.35 | Improved |
| adapt RCAN training pairs to the estimated real degradation with FCA | RCAN · ISO.3 | PSNR: 21.74 -> 25.50 | Improved |
| adapt RCAN training pairs to the estimated real degradation with FCA | RCAN · ISO.[1, 3] | PSNR: 22.51 -> 24.84 | Improved |
| adapt RCAN training pairs to the estimated real degradation with FCA | RCAN · ANI. | PSNR: 22.45 -> 24.93 | Improved |
| add frequency consistency loss to wavelet-discriminator-only training (ablation) | FCA kernel adaptation · ISO.1 | PSNR: 27.07 -> 27.59 | Improved |
| add frequency consistency loss to wavelet-discriminator-only training (ablation) | FCA kernel adaptation · ISO.3 | PSNR: 23.23 -> 25.45 | Improved |
| add frequency consistency loss to wavelet-discriminator-only training (ablation) | FCA kernel adaptation · ISO.[1,3] | PSNR: 23.98 -> 24.63 | Improved |
| add frequency consistency loss to wavelet-discriminator-only training (ablation) | FCA kernel adaptation · ANI. | PSNR: 24.44 -> 24.82 | Improved |
| add wavelet discriminator loss to FDC-only training (ablation) | FCA kernel adaptation · ISO.3 | PSNR: 25.45 -> 25.50 | Improved |
| add wavelet discriminator loss to FDC-only training (ablation) | FCA kernel adaptation · ISO.[1,3] | PSNR: 24.63 -> 24.84 | Improved |
| add wavelet discriminator loss to FDC-only training (ablation) | FCA kernel adaptation · ANI. | PSNR: 24.82 -> 24.93 | Improved |
| add wavelet discriminator loss to FDC-only training (ablation) | FCA kernel adaptation · ISO.1 | PSNR: 27.59 -> 27.35 | Hurt |
Closest papers by meaning
- Learning Omni-Frequency Region-adaptive Representations for Real Image Super-Resolution (AAAI 2021 · similarity 0.91)
- Unsupervised Real-World Super-Resolution: A Domain Adaptation Perspective (ICCV 2021 · similarity 0.90)
- Toward Real-World Single Image Super-Resolution: A New Benchmark and a New Model (ICCV 2019 · similarity 0.89)
- Spectrum-to-Kernel Translation for Accurate Blind Image Super-Resolution (NeurIPS 2021 · similarity 0.89)
- FSR: A General Frequency-Oriented Framework to Accelerate Image Super-resolution Networks (AAAI 2023 · similarity 0.89)
- Learning Controllable Degradation for Real-World Super-Resolution via Constrained Flows (ICML 2023 · similarity 0.89)
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.
