Research map / Video / Video segmentation and tracking
High-Quality and Efficient Turbulence Mitigation with Events
Xiaoran Zhang, Jian Ding, Yuxing Duan, Haoyue Liu, Gang Chen, Yi Chang, Luxin Yan
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 EHETM (events + 5 frames) instead of DATUM | EHETM · CTTH | PSNR: 34.89 -> 35.17 | Improved |
| use EHETM instead of DATUM | EHETM · LATH | NIQE: 10.87 -> 10.12 | Improved |
| use EHETM instead of DATUM | EHETM · system latency | latency (ms): 1560 -> 160 | Improved |
| use EHETM instead of DATUM | EHETM · data size | data size (%): 100.0 -> 22.7 | Improved |
| use EHETM instead of MambaTM | EHETM · CTTH | PSNR: 34.38 -> 35.17 | Improved |
| use multi-frame instead of single-frame methods | multi-frame TM · CTTH | PSNR: 28.17 -> 34.89 | Improved |
| combine RMAB, event tube optimization and EPAW (ablation) | EHETM · CTTH ablation | PSNR: 31.13 -> 35.17 | Improved |
| add event tube optimization to RMAB (ablation) | EHETM · CTTH ablation | PSNR | Improved |
| add event polarity alternation weighting (ablation) | EHETM · CTTH ablation | PSNR | Improved |
| increase inference data from 50 to 100 frames (ablation) | multi-frame TM methods · CTTH / UDET | LPIPS: reduction < 0.05 | No clear effect |
Closest papers by meaning
- Spatio-Temporal Turbulence Mitigation: A Translational Perspective (CVPR 2024 · similarity 0.88)
- Temporally Consistent Atmospheric Turbulence Mitigation with Neural Representations (NeurIPS 2024 · similarity 0.86)
- Turb-Seg-Res: A Segment-then-Restore Pipeline for Dynamic Videos with Atmospheric Turbulence (CVPR 2024 · similarity 0.86)
- Learning Phase Distortion with Selective State Space Models for Video Turbulence Mitigation (CVPR 2025 · similarity 0.85)
- Event-3DGS: Event-based 3D Reconstruction Using 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.84)
- LaSe-E2V: Towards Language-guided Semantic-aware Event-to-Video Reconstruction (NeurIPS 2024 · similarity 0.84)
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.
