Research map / Learning theory and algorithms / Matrix and tensor methods
Learning Sparse and Low-Rank Priors for Image Recovery via Iterative Reweighted Least Squares Minimization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use learned weights with the convex ℓ1 sparse prior instead of the unweighted ℓ1 prior (ablation) | LIRLS grayscale image recovery · Grayscale deblurring benchmarks | PSNR: 31.98 -> 32.70; 35.19 -> 35.75 | Helped |
| Use learned weights with the ℓ1 sparse prior instead of the unweighted ℓ1 prior (ablation) | LIRLS grayscale image recovery · MRI reconstruction | PSNR: 28.53 -> 30.11; 25.86 -> 28.02 | Helped |
| Use a learned nonconvex norm order p<1 rather than the nuclear norm (ablation) | LIRLS color image recovery · Color deblurring | PSNR: 33.09 -> 34.17 | Helped |
| Use learned weights with the S_p^p low-rank prior (ablation) | LIRLS color image recovery · Color deblurring | PSNR: 34.17 -> 34.24 | No clear effect |
| Use learned weights with the S_p^p low-rank prior (ablation) | LIRLS color image recovery · McMaster demosaicking, 3% noise | PSNR: 32.44 -> 33.34 | Helped |
| Use learned weights with the S_p^p low-rank prior (ablation) | LIRLS color image recovery · McMaster demosaicking, 0% noise | PSNR: 37.66 -> 37.62 | No clear effect |
| Use learned weights with the ℓp^p sparse prior (ablation) | LIRLS grayscale image recovery · MRI reconstruction | PSNR: 28.99 -> 29.75; 26.19 -> 27.30 | Hurt |
| Use the convex ℓ1 weighted prior instead of the nonconvex ℓp^p weighted prior (ablation) | LIRLS grayscale image recovery · MRI reconstruction | PSNR: 29.75 -> 30.11; 27.30 -> 28.02 | Helped |
| Use learned weights with the ℓ1 sparse prior (ablation) | LIRLS grayscale image recovery · Grayscale super-resolution x2, 0% noise | PSNR: 28.20 -> 28.76 | Helped |
Closest papers by meaning
- Precise asymptotics of reweighted least-squares algorithms for linear diagonal networks (NeurIPS 2024 · similarity 0.89)
- Iteratively Reweighted Least Squares for Basis Pursuit with Global Linear Convergence Rate (NeurIPS 2021 · similarity 0.89)
- Iterative Regularization with k-support Norm: An Important Complement to Sparse Recovery (AAAI 2024 · similarity 0.87)
- A Restoration Network as an Implicit Prior (ICLR 2024 · similarity 0.87)
- NPN: Non-Linear Projections of the Null-Space for Imaging Inverse Problems (NeurIPS 2025 · similarity 0.87)
- Global Minimizers of $\ell^p$-Regularized Objectives Yield the Sparsest ReLU Neural Networks (NeurIPS 2025 · 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.