Research map / Video / Video segmentation and tracking
Event-based Motion Deblurring with Unpaired Data
Hoonhee Cho, Yuhwan Jeong, Kuk-Jin Yoon
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 EDI-derived self-supervision loss Lself (ablation) | EMP · REBlur (ablation) | PSNR: 28.76 -> 33.92 | Improved |
| add EDI cue as extra input, then frequency modulation module (ablation) | EMP · REBlur (ablation) | PSNR: 28.76 -> 29.24 (cumulative before Lself) | Improved |
| add confidence-weighting to the self-supervision loss and combine all components (ablation) | EMP · REBlur (ablation) | PSNR: 35.30 -> 35.49 | Improved |
| add Part 2 (blur transfer to unpaired sharp images) (ablation) | EMP · REBlur (part ablation) | PSNR: 33.45 -> 34.74 | Improved |
| add Part 3 (ablation) | EMP · REBlur (part ablation) | PSNR: 34.74 -> 35.23 | Improved |
| add Part 4 (reblurring cycle consistency) (ablation) | EMP · REBlur (part ablation) | PSNR: 35.23 -> 35.49 | Improved |
| train EMP unpaired instead of EVDI/Blur2Blur | unpaired baselines · EventAid | PSNR: 31.05 (EVDI) -> 32.99 | Improved |
| use paired supervision with GEM instead of unpaired EMP(GEM) | EMP (GEM) · EventAid | PSNR: 33.24 (paired GEM) vs 32.99 | Hurt |
| train unpaired with EMP vs supervised GEM | EMP (GEM) · HighREV | PSNR: 32.61 (paired GEM) vs 31.16 | Hurt |
| train EMP(GEM) unpaired instead of paired FFTformer | FFTformer · HighREV | PSNR: 30.83 -> 31.16 | Improved |
Closest papers by meaning
- Separation for Better Integration: Disentangling Edge and Motion in Event-based Deblurring (ICCV 2025 · similarity 0.94)
- Frequency-aware Event-based Video Deblurring for Real-World Motion Blur (CVPR 2024 · similarity 0.92)
- Motion-adaptive Transformer for Event-based Image Deblurring (AAAI 2025 · similarity 0.90)
- Learning Scale-Aware Spatio-temporal Implicit Representation for Event-based Motion Deblurring (ICML 2024 · similarity 0.89)
- Asymmetric Hierarchical Difference-aware Interaction Network for Event-guided Motion Deblurring (AAAI 2025 · similarity 0.89)
- Diffusion-based Event Generation for High-Quality Image Deblurring (CVPR 2025 · similarity 0.89)
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.
