Research map / Video / Video segmentation and tracking
Real-Time Video Super-Resolution With Spatio-Temporal Networks and Motion Compensation
Check 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 5-frame early fusion instead of single-frame processing (ablation) | Video ESPCN · CDVL ×3, 9 layers | PSNR: 37.80 -> 37.99 | Helped |
| Use 5-frame slow fusion instead of single-frame processing (ablation) | Video ESPCN · CDVL ×3, 9 layers | PSNR: 37.80 -> 37.99 | Helped |
| Use 5-frame slow fusion instead of early fusion (ablation) | Video ESPCN · CDVL ×3, 9 layers | GOps: 17.22 -> 15.19 | Helped |
| Use 5-frame slow fusion instead of early fusion (ablation) | Video ESPCN · CDVL ×3, 7 layers | PSNR: 37.92 -> 37.83 | Hurt |
| Share weights across temporal positions in slow fusion (ablation) | Video ESPCN · CDVL ×3, 7 layers | PSNR: 37.83 -> 37.74 | Hurt |
| Share weights across temporal positions in slow fusion (ablation) | Video ESPCN · CDVL ×3, 7 layers | GOps: 10.65 -> 8.94 | Helped |
| Add learned motion compensation to 3-frame early fusion (ablation) | Video ESPCN · CDVL ×3 | PSNR: 37.842 -> 37.928; 37.889 -> 37.961; 37.956 -> 38.019; 37.980 -> 38.060 | Helped |
| Use 7 input frames rather than 5 in early fusion (ablation) | Video ESPCN · CDVL ×3 | PSNR | Hurt |
| Use 3-frame early fusion rather than independent single-frame processing (ablation) | Video ESPCN · Vid4 ×3 | PSNR: 26.97 -> 27.05 | Helped |
Closest papers by meaning
- Fast Spatio-Temporal Residual Network for Video Super-Resolution (CVPR 2019 · similarity 0.92)
- Time Without Time: Pseudo-Temporal Representation for Space-Time Super-Resolution (CVPR 2026 · similarity 0.91)
- RSTT: Real-Time Spatial Temporal Transformer for Space-Time Video Super-Resolution (CVPR 2022 · similarity 0.91)
- Residual Invertible Spatio-Temporal Network for Video Super-Resolution (AAAI 2019 · similarity 0.91)
- Look Back and Forth: Video Super-Resolution With Explicit Temporal Difference Modeling (CVPR 2022 · similarity 0.91)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.88)
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.