Research map / Video / Video segmentation and tracking
Joint Video Multi-Frame Interpolation and Deblurring Under Unknown Exposure Time
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the full exposure- and motion-adaptive VIDUE | joint video interpolation and deblurring · GoPro ×8, GoPro ×16, and Adobe ×8 | PSNR: 36.32 vs 35.01; 35.63 vs 34.50; 30.10 vs 28.49; 28.74 vs 27.23; 27.91 vs 26.94 | Helped |
| Add a learned exposure representation using supervised contrastive learning (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 34.79 -> 36.32; 34.13 -> 35.63 | Helped |
| Replace contrastive exposure learning with ordinal regression (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 36.32 -> 35.44; 35.63 -> 34.97 | Hurt |
| Provide the known exposure time rather than infer it (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 36.32 -> 36.41; 35.63 -> 35.87 | Helped |
| Add exposure-adaptive convolution to the exposure-agnostic baseline (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 34.79 -> 36.05; 34.13 -> 34.87 | Helped |
| Add inter-motion analysis to E_Conv-only VIDUE (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 36.05 -> 36.18; 34.87 -> 35.27 | Helped |
| Add intra-motion analysis and motion refinement without inter-motion analysis (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 36.05 -> 36.20; 34.87 -> 35.24 | Helped |
| Add motion refinement to the variant with E_Conv and both motion analyses (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 36.23 -> 36.32; 35.28 -> 35.63 | Mixed |
| Replace exposure-adaptive convolution with squeeze-and-excitation (ablation) | VIDUE · GoPro-5:8 and GoPro-7:8 | PSNR: 36.12 -> 36.32; 35.39 -> 35.63 | Mixed |
Closest papers by meaning
- Event-Based Frame Interpolation With Ad-Hoc Deblurring (CVPR 2023 · similarity 0.91)
- Unifying Motion Deblurring and Frame Interpolation With Events (CVPR 2022 · similarity 0.91)
- Event-Based Blurry Frame Interpolation Under Blind Exposure (CVPR 2023 · similarity 0.91)
- EDEN: Enhanced Diffusion for High-quality Large-motion Video Frame Interpolation (CVPR 2025 · similarity 0.87)
- Motion Blur Decomposition with Cross-shutter Guidance (CVPR 2024 · similarity 0.87)
- Dynamic Gaussian Splatting from Defocused and Motion-blurred Monocular Videos (NeurIPS 2025 · 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.