Research map / Image generation / Image restoration
Efficient Model-Driven Network for Shadow Removal
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 |
|---|---|---|---|
| replace the RB with a DMRB in the initialization network (ablation) | the shadow-removal network · ISTD test, all image | PSNR: 28.37 -> 28.52 | Helped |
| replace the RB with a DMRB in the initialization network (ablation) | the shadow-removal network · ISTD test, all image | RMSE: 5.74 -> 5.50 | Helped |
| add iterative stages (ablation) | the RB initialization network · ISTD test, all image | PSNR: 28.52 -> 29.44 | Helped |
| add iterative stages (ablation) | the RB initialization network · ISTD test, all image | RMSE: 5.50 -> 5.17 | Helped |
| add the transformation-map regularization term (ablation) | the shadow-removal network · ISTD test, all image | PSNR: 29.41 -> 29.85 | Helped |
| add the transformation-map regularization term (ablation) | the shadow-removal network · ISTD test, all image | RMSE: 5.46 -> 5.09 | Helped |
| use 1 iterative stage instead of 0 (ablation) | the shadow-removal network · ISTD test, all image | RMSE: 5.50 -> 5.55 | Hurt |
| increase the number of iterative stages from 0 to 4 (ablation) | the shadow-removal network · ISTD test, all image | PSNR: 28.52 -> 29.85 | Helped |
| increase the number of iterative stages from 0 to 4 (ablation) | the shadow-removal network · ISTD test, all image | RMSE: 5.50 -> 5.09 | Helped |
| increase the number of iterative stages from 4 to 5 (ablation) | the shadow-removal network · ISTD test, all image | PSNR: 29.85 -> 29.70 | Hurt |
| increase the number of iterative stages from 4 to 5 (ablation) | the shadow-removal network · ISTD test, all image | RMSE: 5.09 -> 5.18 | Hurt |
| use the proposed model-driven network instead of DHAN | shadow removal · ISTD test, all image | PSNR: 29.11 -> 29.85 | Helped |
Closest papers by meaning
- Detail-Preserving Latent Diffusion for Stable Shadow Removal (CVPR 2025 · similarity 0.89)
- SoftShadow: Leveraging Soft Masks for Penumbra-Aware Shadow Removal (CVPR 2025 · similarity 0.88)
- When Shadow Removal Meets Intrinsic Image Decomposition: A Joint Learning Framework Using Unpaired Data (AAAI 2025 · similarity 0.88)
- Polarization Guided Mask-Free Shadow Removal (AAAI 2025 · similarity 0.87)
- OmniSR: Shadow Removal Under Direct and Indirect Lighting (AAAI 2025 · similarity 0.87)
- DeS3: Adaptive Attention-Driven Self and Soft Shadow Removal Using ViT Similarity (AAAI 2024 · similarity 0.87)
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.