Research map / Image generation / Fast diffusion sampling
RTE-FM-Dehazer: Radiative Transfer Equation Inspired Flow Matching for Real-World Image Dehazing
Chenfeng Wei, Chun Wang, Boyang Zhao, Si Zuo, Shenhong WANG, Chenguang Yang
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 RTE regularizer to vanilla flow matching (ablation) | RTE-FM-Dehazer · NH-HAZE | PSNR: 13.78 -> 19.53 | Improved |
| add RTE regularizer to vanilla flow matching (ablation) | RTE-FM-Dehazer · SMOKE | PSNR: 13.77 -> 18.67 | Improved |
| add RTE regularizer to vanilla flow matching (ablation) | RTE-FM-Dehazer · NH-HAZE | LPIPS: 0.406 -> 0.233 | Improved |
| add RTE regularizer to vanilla flow matching (ablation) | RTE-FM-Dehazer · RESIDE-6K | PSNR: 17.51 -> 21.77 | Improved |
| add RTE regularizer to vanilla flow matching (ablation) | RTE-FM-Dehazer · P-HAZE | LPIPS: 0.159 -> 0.159 | No clear effect |
| add both diffusion and absorption terms instead of absorb-only (ablation) | RTE-FM-Dehazer · NH-HAZE | PSNR: 12.77 -> 19.53 | Improved |
| add both terms instead of diffusion-only (ablation) | RTE-FM-Dehazer · NH-HAZE | PSNR: 13.85 -> 19.53 | Improved |
| use RTE regularizer instead of dark-channel prior regularizer (ablation) | RTE-FM-Dehazer · NH-HAZE | PSNR: 12.22 -> 19.53 | Improved |
| use RTE-regularized FM instead of rectified flow (ablation) | RTE-FM-Dehazer · SMOKE | PSNR: 13.40 -> 18.67 | Improved |
| use RTE-FM instead of vanilla FM (ablation) | RTE-FM-Dehazer · I-HAZE | PSNR: 18.27 -> 18.81 | Improved |
| RTE-FM-Dehazer vs strongest baseline LHTD | RTE-FM-Dehazer · NH-HAZE | PSNR: 15.60 -> 19.53 | Improved |
| RTE-FM-Dehazer vs DEANet on synthetic benchmark | RTE-FM-Dehazer · RESIDE-6K | PSNR: 31.78 -> 21.77 | Hurt |
Closest papers by meaning
- HazeFlow: Revisit Haze Physical Model as ODE and Non-Homogeneous Haze Generation for Real-World Dehazing (ICCV 2025 · similarity 0.92)
- PRISM: Rethinking Atmospheric Scattering Reconstruction as a Unified Understanding and Restoration Model for Real-world Dehazing (NeurIPS 2026 · similarity 0.91)
- Learning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing (CVPR 2025 · similarity 0.90)
- Depth-Centric Dehazing and Depth-Estimation from Real-World Hazy Driving Video (AAAI 2025 · similarity 0.88)
- PEIE: Physics Embedded Illumination Estimation for Adaptive Dehazing (AAAI 2025 · similarity 0.87)
- Real-world Image Dehazing with Coherence-based Pseudo Labeling and Cooperative Unfolding Network (NeurIPS 2024 · 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.
