Research map / Image generation / Image restoration
Blind Image Deblurring With Local Maximum Gradient Prior
Liang Chen, Faming Fang, Tingting Wang, Guixu Zhang
Check 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 the LMG prior to the gradient-sparse deblurring model (ablation) | LMG blind deblurring · Levin et al. dataset [13] | success percent under error ratio 1.8 | Improved |
| use the LMG prior alone, without the gradient-sparsity term (ablation) | LMG blind deblurring · Levin et al. dataset [13] | error ratio | Hurt |
| constrain the LMG-related term with an L1 norm rather than L0 or L2 (ablation) | LMG blind deblurring · Levin et al. dataset [13] | error ratio | Improved |
| change the LMG patch size among 15 × 15, 25 × 25, 35 × 35, and 45 × 45 (ablation) | LMG blind deblurring · Levin et al. dataset [13] | average PSNR: 33.05, 33.12, 33.29, 33.12 | No clear effect |
| use LMG blind deblurring instead of the strongest reported baseline | blind deblurring on dataset [9] · Dataset [9] | average PSNR: 30.15 -> 30.37 | Improved |
| use LMG blind deblurring instead of Pan et al. [20] | blind deblurring on Levin et al. dataset [13] · Levin et al. dataset [13] | success percent under error ratio 1.8: 93.75% -> 100% | Improved |
| apply the method to the blurred image without first filtering non-Gaussian noise (ablation) | LMG blind deblurring · salt-and-pepper-noisy blurred image | deblurring quality | Hurt |
| use LMG blind deblurring rather than Xu et al. [27] (C++) | blind deblurring · 225 × 225 image | running time: 1.22 -> 65.20 | Hurt |
Closest papers by meaning
- Blind image deblurring with noise-robust kernel estimation (ECCV 2024 · similarity 0.90)
- Blind Deblurring for Saturated Images (CVPR 2021 · similarity 0.90)
- Non-Uniform Blind Deblurring by Reblurring (ICCV 2017 · similarity 0.89)
- Noise-Blind Image Deblurring (CVPR 2017 · similarity 0.89)
- Enhanced Sparse Model for Blind Deblurring (ECCV 2020 · similarity 0.89)
- 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.
