Research map / Video / Video segmentation and tracking
RSTT: Real-Time Spatial Temporal Transformer for Space-Time Video Super-Resolution
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 a unified transformer with reusable encoder dictionaries queried by decoders instead of separate temporal and spatial modules | RSTT · Vid4 | FPS: 26.19 vs 15.59 FPS; 4.49 vs 11.10 parameters (Millions) | Helped |
| Add the spatial reconstruction module (ablation) | RSTT-M and RSTT-S · Vid4 and Vimeo-90K splits | PSNR: RSTT-M-Recon vs RSTT-M: 26.37 vs 26.37 on Vid4; 36.80 vs 36.78 on Vimeo-Fast; 35.66 vs 35.62 on Vimeo-Medium; 33.58 vs 3 | No clear effect |
| Add the spatial reconstruction module (ablation) | RSTT-M and RSTT-S · Vid4 and Vimeo-90K splits | FPS: 17.02 vs 19.07; 22.62 vs 26.19 | Hurt |
| Replace multi-head cross attention with concatenation (ablation) | RSTT-S decoder · Vid4, Vimeo-Fast, Vimeo-Medium, and Vimeo-Slow | PSNR: 26.29 -> 26.18; 36.58 -> 36.29; 35.43 -> 35.20; 33.30 -> 33.15 | Hurt |
| Replace multi-head cross attention with addition (ablation) | RSTT-S decoder · Vid4, Vimeo-Fast, Vimeo-Medium, and Vimeo-Slow | PSNR: 26.29 -> 26.13; 36.58 -> 36.25; 35.43 -> 35.15; 33.30 -> 33.07 | Hurt |
Closest papers by meaning
- Time Without Time: Pseudo-Temporal Representation for Space-Time Super-Resolution (CVPR 2026 · similarity 0.92)
- Temporal Modulation Network for Controllable Space-Time Video Super-Resolution (CVPR 2021 · similarity 0.92)
- Real-Time Video Super-Resolution With Spatio-Temporal Networks and Motion Compensation (CVPR 2017 · similarity 0.91)
- BF-STVSR: B-Splines and Fourier---Best Friends for High Fidelity Spatial-Temporal Video Super-Resolution (CVPR 2025 · similarity 0.90)
- EvSTVSR: Event Guided Space-Time Video Super-Resolution (AAAI 2025 · similarity 0.88)
- EvEnhancer: Empowering Effectiveness, Efficiency and Generalizability for Continuous Space-Time Video Super-Resolution with Events (CVPR 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.