Research map / Learning theory and algorithms / Optimal transport
When Schrodinger Bridge Meets Real-World Image Dehazing with Unpaired Training
Yunwei Lan, Zhigao Cui, Xin Luo, Chang Liu, Nian Wang, Menglin Zhang, Yanzhao Su, Dong Liu
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 prompt learning (haze-aware prompt) (ablation) | DehazeSB backbone · Haze2020 | FID: 80.260 -> 73.968 | Improved |
| add prompt learning (haze-aware prompt) (ablation) | DehazeSB backbone · OHAZE | PSNR: 17.410 -> 17.445 | No clear effect |
| add detail-preserving regularization (ablation) | DehazeSB backbone · Haze2020 | FID: 80.260 -> 75.192 | Improved |
| add detail-preserving regularization (ablation) | DehazeSB backbone · OHAZE | PSNR: 17.410 -> 18.687 | Improved |
| combine prompt learning and detail-preserving regularization (ablation) | DehazeSB backbone · Haze2020 | FID: 80.260 -> 69.796 | Improved |
| add physical prior regularization to the backbone with PL (ablation) | DehazeSB · OHAZE | PSNR: 17.445 -> 18.085 | Improved |
| add high-frequency detail regularization to the backbone with PL (ablation) | DehazeSB · OHAZE | PSNR: 17.445 -> 17.534 | Improved |
| add physical prior regularization (ablation) | DehazeSB · Haze2020 | NIQE: 3.690 -> 3.753 | Hurt |
| remove Markovian discriminator (ablation) | DehazeSB · qualitative dehazing | haze removal | Hurt |
| Schrodinger Bridge unpaired dehazing vs UNSB | RTTS · RTTS | FID: 56.668 -> 53.120 | Improved |
| DehazeSB vs KANet on NIQE | Haze2020 · Haze2020 | NIQE: 3.742 -> 3.743 | Hurt |
Closest papers by meaning
- Exploiting Diffusion Prior for Real-World Image Dehazing with Unpaired Training (AAAI 2025 · similarity 0.90)
- Learning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing (CVPR 2025 · similarity 0.89)
- Advancing Real-World Image Dehazing: Perspective, Modules, and Training (TPAMI 2024 · similarity 0.88)
- PromptHaze: Prompting Real-world Dehazing via Depth Anything Model (AAAI 2025 · similarity 0.88)
- Dehaze-RetinexGAN: Real-World Image Dehazing via Retinex-based Generative Adversarial Network (AAAI 2025 · similarity 0.87)
- Prior-guided Hierarchical Harmonization Network for Efficient Image Dehazing (AAAI 2025 · similarity 0.86)
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.
