Research map / Video / Video segmentation and tracking
Space-Time-Aware Multi-Resolution Video Enhancement
Muhammad Haris, Greg Shakhnarovich, Norimichi Ukita
Check 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 |
|---|---|---|---|
| Joint STAR-ST space-time super-resolution instead of the strongest combined RBPN → DAIN baseline | STAR · Vimeo90K ST-SR | PSNR: 30.623 -> 30.830 | Improved |
| Fine-tune STAR using the task-specific Space and Space-Time losses (ablation) | STAR · Vimeo90K input frames | PSNR: 31.601 -> 31.883 | Improved |
| Fine-tune STAR using Space loss (ablation) | STAR · Vimeo90K input frames | PSNR: 31.601 -> 32.026 | Improved |
| Add refinement Stage 2 (ablation) | STAR · Vimeo90K input frames | PSNR: 30.920 -> 31.920 | Improved |
| Add refinement Stage 2 (ablation) | STAR · Vimeo90K in-between frames | PSNR: 30.002 -> 30.365 | Improved |
| Use flow images and the motion network (ablation) | STAR · Vimeo90K input frames | PSNR: 31.489 -> 31.920 | Improved |
| Use flow images and the motion network (ablation) | STAR · Vimeo90K in-between frames | PSNR: 30.086 -> 30.365 | Improved |
| Add Flow Refinement (ablation) | STAR · Vimeo90K in-between frames | PSNR: 30.229 -> 30.365 | Improved |
| Use RBPN instead of DBPN as the spatial SR module (ablation) | STAR-ST · Vimeo90K in-between frames | PSNR: 30.540 -> 30.704 | Improved |
| Use STAR-T rather than STAR-ST for the second 2× temporal upsampling stage (ablation) | STAR-ST upscaling path · Vimeo90K setuplet test, in-between frames | PSNR: 27.186 -> 27.640 | Improved |
| Add the T-SR path on S-HR (ablation) | ST-SR · Vimeo90K in-between frames | PSNR: 30.704 -> 30.712 | Improved |
| Add the T-SR path on S-LR (ablation) | ST-SR · Vimeo90K in-between frames | PSNR: 30.704 -> 30.760 | Improved |
Closest papers by meaning
- Time Without Time: Pseudo-Temporal Representation for Space-Time Super-Resolution (CVPR 2026 · similarity 0.92)
- Spatial-Temporal Space Hand-in-Hand: Spatial-Temporal Video Super-Resolution via Cycle-Projected Mutual Learning (CVPR 2022 · similarity 0.91)
- Look Back and Forth: Video Super-Resolution With Explicit Temporal Difference Modeling (CVPR 2022 · similarity 0.91)
- RSTT: Real-Time Spatial Temporal Transformer for Space-Time Video Super-Resolution (CVPR 2022 · similarity 0.91)
- EvSTVSR: Event Guided Space-Time Video Super-Resolution (AAAI 2025 · similarity 0.89)
- BF-STVSR: B-Splines and Fourier---Best Friends for High Fidelity Spatial-Temporal Video Super-Resolution (CVPR 2025 · 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.
