Research map / Image generation / Image restoration
Fast Single Image Reflection Suppression via Convex Optimization
Check 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 the convex gradient-thresholding suppression model instead of Arvanitopoulos et al. | Single-image reflection suppression · Synthetic images (four cases) | PSNR: 20.87 -> 20.99; 16.80 -> 16.93; 19.42 -> 19.44; 15.10 -> 15.14 | Helped |
| Use the convex gradient-thresholding suppression model instead of Arvanitopoulos et al. | Single-image reflection suppression · Synthetic images (four cases) | SSIM: 0.896 -> 0.903; 0.716 -> 0.736; 0.896 -> 0.897; 0.787 -> 0.789 | Helped |
| Use the proposed direct convex solution instead of Li and Brown | Single-image reflection suppression · Synthetic images (four cases), 512 × 512 | execution time (sec): 12.06 -> 0.19; 11.68 -> 0.18; 7.25 -> 0.19; 7.69 -> 0.19 | Helped |
| Use the proposed direct convex solution instead of Li and Brown | Single-image reflection suppression · Real-world images (four cases), 1080 × 1440 | execution time (sec): 39.06 -> 1.46; 52.60 -> 1.36; 17.90 -> 1.40; 10.75 -> 1.15 | Helped |
| Increase gradient threshold h (ablation) | Proposed reflection suppression model · Input images | reflection suppression | Mixed |
| Use a positive epsilon fidelity term to ensure uniqueness (ablation) | Proposed convex model · Single-image reflection suppression | solution uniqueness | Helped |
| Use the proposed method when reflection edges are sharp and strong | Proposed reflection suppression model · Failure cases with sharp reflections | reflection removal | Hurt |
Closest papers by meaning
- Single Image Reflection Suppression (CVPR 2017 · similarity 0.93)
- Anti-Glare: Tightly Constrained Optimization for Eyeglass Reflection Removal (CVPR 2017 · similarity 0.89)
- Revisiting Single Image Reflection Removal In the Wild (CVPR 2024 · similarity 0.84)
- Removing Reflections from RAW Photos (CVPR 2025 · similarity 0.83)
- PolarFree: Polarization-based Reflection-Free Imaging (CVPR 2025 · similarity 0.83)
- Flash-Split: 2D Reflection Removal with Flash Cues and Latent Diffusion Separation (CVPR 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.