Research map / Image generation / Image restoration
Single Image Highlight Removal with a Sparse and Low-Rank Reflection Model
Jie Guo, Zuojian Zhou, Limin Wang
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 |
|---|---|---|---|
| use accurate rather than estimated illumination chromaticity (ablation) | SLRR highlight removal · Highlight removal, Masks | PSNR: 31.5 -> 34.4 | Improved |
| use accurate rather than estimated illumination chromaticity (ablation) | SLRR highlight removal · Highlight removal, Dragon | PSNR: 30.5 -> 31.8 | Improved |
| use accurate rather than estimated illumination chromaticity (ablation) | SLRR highlight removal · Highlight removal, Fruit | PSNR: 36.5 -> 36.4 | No clear effect |
| use accurate rather than estimated illumination chromaticity (ablation) | SLRR highlight removal · Rendered scene (Fig. 2 left) | PSNR: 33.0 -> 37.1 | Improved |
| increase color dictionary size K from 1 to 10 (ablation) | SLRR highlight removal · Real images with ground truth (Cups, Animals, Masks, Fruit) | PSNR and SSIM | Improved |
| increase color dictionary size K from 10 to 80 (ablation) | SLRR highlight removal · Real images with ground truth (Cups, Animals, Masks, Fruit) | PSNR and SSIM | No clear effect |
| use slack variables for non-negativity instead of clamping negative entries to zero (ablation) | SLRR ALM solver · Animals and Fruit | convergence speed | Improved |
Closest papers by meaning
- Towards High-Quality Specular Highlight Removal by Leveraging Large-Scale Synthetic Data (ICCV 2023 · similarity 0.90)
- One-Step Specular Highlight Removal with Adapted Diffusion Models (ICCV 2025 · similarity 0.90)
- A Multi-Task Network for Joint Specular Highlight Detection and Removal (CVPR 2021 · similarity 0.88)
- SHDocs: A dataset, benchmark, and method to efficiently generate high-quality, real-world specular highlight data with near-perfect alignment (NeurIPS 2024 · similarity 0.86)
- Hierarchical Adaptive Filtering Network for Text Image Specular Highlight Removal (CVPR 2025 · similarity 0.85)
- PHR-DIFF: Portrait Highlights Removal via Patch-aware Diffusion Model (AAAI 2025 · similarity 0.84)
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.
