Research map / Image generation / Image restoration
Single Image Defocus Deblurring via Implicit Neural Inverse Kernels
Yuhui Quan, Xin Yao, Hui Ji
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 |
|---|---|---|---|
| share INR atoms across scales via multi-scale dictionary structure (ablation) | INIKNet · RealDOF (trained on DPDD) | PSNR (dB): 24.916 -> 25.231 | Improved |
| parameterize atoms with INR instead of plain upsampling (ablation) | INIKNet · RealDOF (trained on DPDD) | PSNR (dB): 25.058 -> 25.231 | Improved |
| use bi-LSTM duplex scale-recurrence instead of unidirectional (ablation) | INIKNet · RealDOF (trained on DPDD) | PSNR (dB): 25.099 -> 25.231 | Improved |
| multi-scale INR dictionary versus non-multi-scale (ablation) | INIKNet · RTF (trained on DPDD) | PSNR (dB): 25.136 -> 25.450 | Improved |
| INR atoms versus plain atoms (ablation) | INIKNet · RTF (trained on DPDD) | SSIM: 0.819 -> 0.834 | Improved |
| INIKNet versus Restormer on DPDD | INIKNet · DPDD | SSIM: 0.811 -> 0.803 | Hurt |
| INIKNet versus AIFNet on RTF (LFDOF-trained) | INIKNet · RTF (trained on LFDOF) | PSNR (dB): 27.552 -> 27.401 | Hurt |
| compact INIKNet versus much larger models | INIKNet · DPDD | PSNR (dB): 25.980 -> 26.055 | 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.
- Restormer — Restormer: Efficient Transformer for High-Resolution Image Restoration (CVPR 2022) · better on 3 of 5 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
- Prior and Prediction Inverse Kernel Transformer for Single Image Defocus Deblurring (AAAI 2024 · similarity 0.94)
- Neumann Network With Recursive Kernels for Single Image Defocus Deblurring (CVPR 2023 · similarity 0.94)
- Single Image Defocus Deblurring Using Kernel-Sharing Parallel Atrous Convolutions (ICCV 2021 · similarity 0.92)
- Residual Diffusion Deblurring Model for Single Image Defocus Deblurring (AAAI 2025 · similarity 0.92)
- Learnable Blur Kernel for Single-Image Defocus Deblurring in the Wild (AAAI 2023 · similarity 0.92)
- DeblurDiff: Real-Word Image Deblurring with Generative Diffusion Models (NeurIPS 2025 · similarity 0.88)
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.
