Research map / Image generation / Image restoration
Hybrid CNN-Transformer Feature Fusion for Single Image Deraining
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 |
|---|---|---|---|
| use HCT-FFN instead of IPT | single-image deraining · Rain100L | PSNR: 41.62 -> 39.70 | Hurt |
| use HCT-FFN instead of IPT | single-image deraining · Rain100L | SSIM: 0.988 -> 0.985 | Hurt |
| use HCT-FFN instead of Uformer-B | single-image deraining · Rain100H | PSNR: 31.06 -> 31.51 | Helped |
| use HCT-FFN instead of Uformer-B | single-image deraining · Rain100H | SSIM: 0.908 -> 0.910 | Helped |
| use HCT-FFN instead of MPRNet | single-image deraining · Rain12 | PSNR: 37.26 -> 37.54 | Helped |
| use HCT-FFN instead of PReNet | single-image deraining · Rain12 | SSIM: 0.961 -> 0.963 | Helped |
| use HCT-FFN instead of MPRNet | single-image deraining · RainDS-RS100 | PSNR: 26.86 -> 27.02 | Helped |
| use HCT-FFN instead of Uformer-B | single-image deraining · RainDS-RS100 | SSIM: 0.728 -> 0.734 | Helped |
| use HCT-FFN instead of Uformer | single-image deraining · Internet-Data | NIQE: 5.034 -> 4.743 | Helped |
| use HCT-FFN instead of Uformer | single-image deraining · Internet-Data | BRISQUE: 33.293 -> 28.709 | Helped |
| add a second recursive stage rather than use one stage (ablation) | HCT-FFN · Rain100L | PSNR: 36.31 -> 38.87 | Helped |
| add a third recursive stage rather than use two stages (ablation) | HCT-FFN · Rain100L | PSNR: 38.87 -> 39.70 | Helped |
Closest papers by meaning
- Multi-Scale Progressive Fusion Network for Single Image Deraining (CVPR 2020 · similarity 0.92)
- Rethinking Multi-Scale Representations in Deep Deraining Transformer (AAAI 2024 · similarity 0.91)
- Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining (CVPR 2024 · similarity 0.91)
- A Model-Driven Deep Neural Network for Single Image Rain Removal (CVPR 2020 · similarity 0.90)
- Boosting Image De-Raining via Central-Surrounding Synergistic Convolution (AAAI 2025 · similarity 0.89)
- Harnessing Joint Rain-/Detail-aware Representations to Eliminate Intricate Rains (ICLR 2024 · similarity 0.89)
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.