Research map / Image generation / GANs
RIS-GAN: Explore Residual and Illumination with Generative Adversarial Networks for Shadow Removal
Ling Zhang, Chengjiang Long, Xiaolong Zhang, Chunxia Xiao
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 |
|---|---|---|---|
| Add residual and illumination estimates with refinement | RIS-GAN · SRD | RMSE (entire image): 7.83 -> 6.78 | Improved |
| Add residual and illumination estimates with refinement | RIS-GAN · ISTD | RMSE (entire image): 7.10 -> 6.95 | Improved |
| Add the residual branch to the removal and illumination refinement (ablation) | RIS-GAN · SRD | RMSE (entire image): 7.16 -> 6.78 | Improved |
| Add the illumination branch to the removal and residual refinement (ablation) | RIS-GAN · SRD | RMSE (entire image): 7.35 -> 6.78 | Improved |
| Add adversarial loss (ablation) | RIS-GAN · SRD | RMSE (entire image): 7.32 -> 6.78 | Improved |
| Add cross loss (ablation) | RIS-GAN · SRD | RMSE (entire image): 7.27 -> 6.78 | Improved |
| Use only the residual generator rather than the full framework (ablation) | RIS-GAN · SRD | RMSE (entire image): 6.78 -> 8.37 | Hurt |
| Use only the removal generator rather than the full framework (ablation) | RIS-GAN · SRD | RMSE (entire image): 6.78 -> 8.85 | Hurt |
| Use only the illumination generator rather than the full framework (ablation) | RIS-GAN · SRD | RMSE (entire image): 6.78 -> 11.55 | Hurt |
| Use only the residual generator rather than the full framework (ablation) | RIS-GAN · ISTD | RMSE (entire image): 6.95 -> 8.24 | Hurt |
| Use only the removal generator rather than the full framework (ablation) | RIS-GAN · ISTD | RMSE (entire image): 6.95 -> 11.27 | Hurt |
| Use only the illumination generator rather than the full framework (ablation) | RIS-GAN · ISTD | RMSE (entire image): 6.95 -> 15.31 | Hurt |
Closest papers by meaning
- ARGAN: Attentive Recurrent Generative Adversarial Network for Shadow Detection and Removal (ICCV 2019 · similarity 0.92)
- Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal (CVPR 2018 · similarity 0.90)
- Towards Ghost-Free Shadow Removal via Dual Hierarchical Aggregation Network and Shadow Matting GAN (AAAI 2020 · similarity 0.90)
- OmniSR: Shadow Removal Under Direct and Indirect Lighting (AAAI 2025 · similarity 0.88)
- When Shadow Removal Meets Intrinsic Image Decomposition: A Joint Learning Framework Using Unpaired Data (AAAI 2025 · similarity 0.87)
- Polarization Guided Mask-Free Shadow Removal (AAAI 2025 · 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.
