Research map / Image generation / Image restoration
Extreme-Quality Computational Imaging via Degradation Framework
Shiqi Chen, Huajun Feng, Keming Gao, Zhihai Xu, Yueting Chen
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 estimated degradation kernels from the framework instead of Zemax PSFs (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | PSNR: 33.52 -> 35.93 | Improved |
| add unprocess and process operations (energy-domain degradation) (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | PSNR: 35.11 -> 35.93 | Improved |
| add optical geometric priors (PSF size and symmetry constraints) (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | PSNR: 33.96 -> 35.93 | Improved |
| add FOV Encoder (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | PSNR: 34.69 -> 35.93 | Improved |
| add Deformable Decoder (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | PSNR: 35.34 -> 35.93 | Improved |
| add KPN Block with dilated kernels (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | PSNR: 35.26 -> 35.93 | Improved |
| add optical geometric priors (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | LPIPS: 0.2285 -> 0.1223 | Improved |
| add estimated degradation kernels instead of Zemax PSFs (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | LPIPS: 0.2749 -> 0.1223 | Improved |
| add FOV Encoder (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | LPIPS: 0.1622 -> 0.1223 | Improved |
| add Deformable Decoder (ablation) | FOV-KPN with degradation framework · Synthetic test set (DIV2K 50 images, degradation framework) | LPIPS: 0.1245 -> 0.1223 | No clear effect |
| use FOV-KPN instead of KPN (k=19) | KPN · Synthetic test set (DIV2K 50 images, degradation framework) | PSNR: 35.62 -> 35.93 | Improved |
| use FOV-KPN instead of KPN (k=19) | KPN · Synthetic test set (DIV2K 50 images, degradation framework) | NIQE: 1.9759 -> 2.0578 | Hurt |
Closest papers by meaning
- A Physics-Informed Blur Learning Framework for Imaging Systems (CVPR 2025 · similarity 0.86)
- Unsupervised Diffusion-Based Degradation Modeling for Real-World Super-Resolution (AAAI 2025 · similarity 0.84)
- Real-World Mobile Image Denoising Dataset with Efficient Baselines (CVPR 2024 · similarity 0.84)
- A Physics-informed Low-rank Deep Neural Network for Blind and Universal Lens Aberration Correction (CVPR 2024 · similarity 0.84)
- Continuous Optical Zooming: A Benchmark for Arbitrary-Scale Image Super-Resolution in Real World (CVPR 2024 · similarity 0.83)
- Towards Realistic Data Generation for Real-World Super-Resolution (ICLR 2025 · similarity 0.83)
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.
