Research map / Video / Video segmentation and tracking
Look Back and Forth: Video Super-Resolution With Explicit Temporal Difference Modeling
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 |
|---|---|---|---|
| Add region decomposition into low- and high-variance regions (ablation) | ETDM · Vid4 | PSNR (dB): 27.89 -> 28.04; +0.15dB | Helped |
| Add region decomposition into low- and high-variance regions (ablation) | ETDM · UDM10 | PSNR (dB): 39.46 -> 39.68; +0.22dB | Helped |
| Add forward HR-space temporal-difference refinement (ablation) | ETDM with region decomposition · Vid4 | PSNR (dB): 28.04 -> 28.29; +0.25dB | Helped |
| Add forward HR-space temporal-difference refinement (ablation) | ETDM with region decomposition · UDM10 | PSNR (dB): 39.68 -> 39.88; +0.20dB | Helped |
| Add backward refinement using future SR results (ablation) | ETDM with forward refinement · Vid4 | PSNR (dB): 28.29 -> 28.81; +0.52dB | Helped |
| Add backward refinement using future SR results (ablation) | ETDM with forward refinement · UDM10 | PSNR (dB): 39.88 -> 40.11; +0.23dB | Helped |
| Use HR temporal differences instead of optical flow for refinement at N=1 (ablation) | ETDM · Vid4 | PSNR (dB): 28.29 vs 28.18 | Hurt |
| Use HR temporal differences instead of optical flow for refinement at N=1 (ablation) | ETDM · UDM10 | PSNR (dB): 39.76 vs 39.77 | No clear effect |
| Use HR temporal differences instead of optical flow for refinement at N=3 (ablation) | ETDM · Vid4 | PSNR (dB): 28.54 vs 28.81 | Helped |
| Use HR temporal differences instead of optical flow for refinement at N=3 (ablation) | ETDM · UDM10 | PSNR (dB): 40.01 vs 40.11 | Helped |
Closest papers by meaning
- Spatial-Temporal Space Hand-in-Hand: Spatial-Temporal Video Super-Resolution via Cycle-Projected Mutual Learning (CVPR 2022 · similarity 0.91)
- Time Without Time: Pseudo-Temporal Representation for Space-Time Super-Resolution (CVPR 2026 · similarity 0.91)
- Real-Time Video Super-Resolution With Spatio-Temporal Networks and Motion Compensation (CVPR 2017 · similarity 0.91)
- Sequence Matters: Harnessing Video Models in 3D 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 0.89)
- VideoGigaGAN: Towards Detail-rich Video Super-Resolution (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.