Research map / Video / Video segmentation and tracking
Rethinking Video Rain Streak Removal: A New Synthesis Model and a Deraining Network with Video Rain Prior
Shuai Wang, Lei Zhu, Huazhu Fu, Jing Qin, Carola-Bibiane Schönlieb, Wei Feng, Song 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 |
|---|---|---|---|
| add DT module (disentangle rain, motion, background) (ablation) | basic network · RainMotion | PSNR: 30.89 -> 31.22 | Improved |
| add DT module (disentangle rain, motion, background) (ablation) | basic network · RainMotion | SSIM: 0.9351 -> 0.9367 | Improved |
| add AR modules (attentive aggregation) (ablation) | basic+DT · RainMotion | PSNR: 31.22 -> 31.82 | Improved |
| add AR modules (attentive aggregation) (ablation) | basic+DT · RainMotion | SSIM: 0.9367 -> 0.9423 | Improved |
| add rain streak loss LS (ablation) | RDD-Net · RainMotion | PSNR: 31.09 -> 31.82 | Improved |
| add rain streak loss LS (ablation) | RDD-Net · RainMotion | SSIM: 0.9372 -> 0.9423 | Improved |
| add rain motion loss LM (ablation) | RDD-Net · RainMotion | PSNR: 30.92 -> 31.82 | Improved |
| add rain motion loss LM (ablation) | RDD-Net · RainMotion | SSIM: 0.9364 -> 0.9423 | Improved |
| add both LS and LM supervision (ablation) | RDD-Net · RainMotion | PSNR: 30.68 -> 31.82 | Improved |
| add both LS and LM supervision (ablation) | RDD-Net · RainMotion | SSIM: 0.9353 -> 0.9423 | Improved |
| use RDD-Net instead of best prior video methods (FCDN, S2VD) | RDD-Net · RainSynHeavy25 | PSNR: PSNR improvement of 16.85% | Improved |
| use RDD-Net instead of best prior video methods (FCDN, S2VD) | RDD-Net · NTURain | PSNR: PSNR improvement of 0.91% and a SSIM improvement of 0.38% | Improved |
Closest papers by meaning
- Self-Learning Video Rain Streak Removal: When Cyclic Consistency Meets Temporal Correspondence (CVPR 2020 · similarity 0.93)
- Recurrent Multi-Frame Deraining: Combining Physics Guidance and Adversarial Learning (TPAMI 2021 · similarity 0.93)
- Learning to Remove Rain in Video With Self-Supervision (TPAMI 2022 · similarity 0.93)
- Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Deraining (CVPR 2025 · similarity 0.90)
- NightRain: Nighttime Video Deraining via Adaptive-Rain-Removal and Adaptive-Correction (AAAI 2024 · similarity 0.89)
- Boosting Image De-Raining via Central-Surrounding Synergistic Convolution (AAAI 2025 · 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.
