Research map / Image generation / Image restoration
Shadow Removal via Shadow Image Decomposition
Hieu Le, Dimitris Samaras
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 SP-Net with binary mask instead of Gong et al. | state-of-the-art shadow removal · ISTD | RMSE shadow area: 13.3 -> 9.5 | Improved |
| add M-Net shadow matte (ablation) | SP-Net · ISTD | RMSE shadow area: 9.5 -> 7.9 | Improved |
| add M-Net shadow matte (ablation) | SP-Net · ISTD | RMSE all: 4.1 -> 3.9 | Improved |
| add M-Net shadow matte (ablation) | SP-Net · ISTD | RMSE non-shadow area: 3.2 -> 3.1 | No clear effect |
| train on augmented ISTD with scaled shadow parameters | SP+M-Net · ISTD | RMSE shadow area: 7.9 -> 7.4 | Improved |
| train on augmented ISTD with scaled shadow parameters | SP-Net · ISTD | RMSE shadow area: 9.5 -> 9.0 | Improved |
| train on augmented ISTD with scaled shadow parameters | SP+M-Net · ISTD | RMSE all: 3.9 -> 3.8 | Improved |
| jointly train a shadow detector with SP-Net and M-Net (ablation) | SP+M-Net · ISTD | RMSE shadow area: 7.9 -> 8.4 | Hurt |
| remove shadow mask input to SP-Net and M-Net (ablation) | SP+M-Net · ISTD | RMSE shadow area: 7.9 -> 8.3 | Hurt |
| remove shadow mask input to SP-Net and M-Net (ablation) | SP+M-Net · ISTD | RMSE non-shadow area: 3.1 -> 4.9 | Hurt |
| apply linear-regression color correction to ground-truth shadow-free images | ISTD test ground truth · ISTD | RMSE non-shadow area between shadow and shadow-free images: 6.83 -> 2.6 | Improved |
Closest papers by meaning
- Physics-Based Shadow Image Decomposition for Shadow Removal (TPAMI 2021 · similarity 0.97)
- Efficient Model-Driven Network for Shadow Removal (AAAI 2022 · similarity 0.94)
- SoftShadow: Leveraging Soft Masks for Penumbra-Aware Shadow Removal (CVPR 2025 · similarity 0.90)
- When Shadow Removal Meets Intrinsic Image Decomposition: A Joint Learning Framework Using Unpaired Data (AAAI 2025 · similarity 0.90)
- Polarization Guided Mask-Free Shadow Removal (AAAI 2025 · similarity 0.88)
- Detail-Preserving Latent Diffusion for Stable Shadow Removal (CVPR 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.
