Research map / Image generation / Image restoration
Physics-Driven Turbulence Image Restoration with Stochastic Refinement
Ajay Jaiswal, Xingguang Zhang, Stanley H. Chan, Zhangyang 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 |
|---|---|---|---|
| use PiRN with differentiable physics simulator consistency loss instead of TurbNet | turbulence restoration · PiRN-Syn test | PSNR: 23.72 -> 25.40 | Improved |
| add stochastic refinement (conditional DDPM, 10-20 steps) on top of PiRN (ablation) | PiRN · PiRN-Syn test | NIQE: 6.5852 -> 5.5847 | Improved |
| add stochastic refinement on top of PiRN (ablation) | PiRN · PiRN-Syn test | PSNR: 25.40 -> 25.61 | No clear effect |
| PiRN vs TurbNet under strong turbulence | turbulence restoration · PiRN-Syn strong turbulence | PSNR: 23.90 -> 25.02 | Improved |
| restore text images with PiRN before OCR | OCR on Turbulence Text · Turbulence Text | AWDR: 0.623 -> 0.776 | Improved |
| restore video with PiRN-SR before MTCNN face detection | face detection on turbulence video · 400m/500m videos | F1-score: 0.4331 -> 0.4648 (500m) | Improved |
| condition the diffusion model on PiRN output instead of the degraded image (ablation) | stochastic refinement · diffusion training | training stability | Improved |
| use stochastic AT-DDPM restoration instead of deterministic SwinIR before face detection | face detection · 500m video | F1-score: 0.4006 (SwinIR) -> 0.3997 (AT-DDPM) | Hurt |
Closest papers by meaning
- Spatio-Temporal Turbulence Mitigation: A Translational Perspective (CVPR 2024 · similarity 0.86)
- Learning Phase Distortion with Selective State Space Models for Video Turbulence Mitigation (CVPR 2025 · similarity 0.85)
- Turb-Seg-Res: A Segment-then-Restore Pipeline for Dynamic Videos with Atmospheric Turbulence (CVPR 2024 · similarity 0.84)
- NB-GTR: Narrow-Band Guided Turbulence Removal (CVPR 2024 · similarity 0.83)
- Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start (NeurIPS 2025 · similarity 0.83)
- Temporally Consistent Atmospheric Turbulence Mitigation with Neural Representations (NeurIPS 2024 · similarity 0.83)
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.
