Research map / Image generation / Image restoration
Unified Removal of Raindrops and Reflections: A New Benchmark and A Novel Pipeline
Xingyu Liu, Zewei He, Yu Chen, Chunyu Zhu, Zixuan Chen, Xing Luo, Zheming Lu
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 Modulate&Gate module instead of channel concatenation of I_lq and I_s (ablation) | DiffUR3 · RDRF-testing | PSNR: 28.60 -> 29.41 | Improved |
| add Modulate&Gate module instead of channel concatenation of I_lq and I_s (ablation) | DiffUR3 · RDRF-testing | LPIPS: 0.0899 -> 0.0813 | Improved |
| add Modulate block (ablation) | DiffUR3 without Modulate · RDRF-testing | PSNR: 29.11 -> 29.41 | Improved |
| add Gate block (ablation) | DiffUR3 without Gate · RDRF-testing | PSNR: 29.12 -> 29.41 | Improved |
| use both I_lq and I_s as conditions rather than I_lq only (ablation) | DiffUR3 conditioning · RDRF-testing | PSNR: 28.33 -> 29.41 | Improved |
| use both I_lq and I_s as conditions rather than I_s only (ablation) | DiffUR3 conditioning · RDRF-testing | PSNR: 28.54 -> 29.41 | Improved |
| add Fidelity Encoder to VAE decoding (ablation) | DiffUR3 · RDRF-testing | PSNR: 26.95 -> 29.41 | Improved |
| add Fidelity Encoder to VAE decoding (ablation) | DiffUR3 · RDRF-testing | SSIM: 0.8042 -> 0.9372 | Improved |
| add Fidelity Encoder to VAE decoding (ablation) | DiffUR3 · RDRF-testing | LPIPS: 0.1082 -> 0.0813 | Improved |
| replace CFW (w=1.0) with Fidelity Encoder (ablation) | DiffUR3 · RDRF-testing | PSNR: 29.18 -> 29.41 | Improved |
| replace CFW (w=0.5) with Fidelity Encoder (ablation) | DiffUR3 · RDRF-testing | PSNR: 28.78 -> 29.41 | Improved |
| add stage II diffusion generation on top of stage I restoration | DRSformer stage I · RDRF-testing | LPIPS: 0.0925 -> 0.0813 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- UniRestore — UniRestore: Unified Perceptual and Task-Oriented Image Restoration Model Using Diffusion Prior (CVPR 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Raindrop Clarity: A Dual-Focused Dataset for Day and Night Raindrop Removal (ECCV 2024 · similarity 0.90)
- Semantic Segmentation on Raindrop Degraded Images Using Two-Stage Dual Teacher-Student Learning (AAAI 2025 · similarity 0.90)
- Dual-Pixel Raindrop Removal (TPAMI 2024 · similarity 0.88)
- Removing Raindrops and Rain Streaks in One Go (CVPR 2021 · similarity 0.88)
- Joint Bi-Layer Optimization for Single-Image Rain Streak Removal (ICCV 2017 · similarity 0.88)
- Towards Universal Rainy Image Restoration: Benchmark and Baseline (AAAI 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.
