Research map / Video / Video segmentation and tracking
Across Scales & Across Dimensions: Temporal Super-Resolution using Deep Internal Learning
Liad Pollak Zuckerman, Eyal Naor, George Pisha, Shai Bagon, Michal Irani
Read the paperCheck 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 |
|---|---|---|---|
| use across-dimension examples instead of within-dimension examples only (ablation) | Deep internal TSR network · TSR x2 coarse scale | PSNR [dB]: 33.96 -> 34.25 | Improved |
| use across-dimension examples instead of within-dimension examples only (ablation) | Deep internal TSR network · TSR x2 coarse scale | LPIPS: 0.035 -> 0.033 | Improved |
| select the best training configuration per video (ablation) | Deep internal TSR network · TSR x2 coarse scale | PSNR [dB]: 33.96 -> 34.33 | Improved |
| select the best training configuration per video (ablation) | Deep internal TSR network · TSR x2 coarse scale | SSIM: 0.962 -> 0.965 | Improved |
| use unsupervised internal training instead of supervised Flawless | Deep internal TSR network · Challenging videos | PSNR [dB]: 27.58 -> 28.58 | Improved |
| use unsupervised internal training instead of supervised Flawless | Deep internal TSR network · Entire dataset | PSNR [dB]: 28.22 -> 28.63 | Improved |
| use unsupervised internal training instead of supervised Flawless | Deep internal TSR network · Entire dataset | SSIM: 0.918 -> 0.917 | No clear effect |
| use dedicated TSR instead of frame interpolation (DAIN) | Deep internal TSR network · Entire dataset | PSNR [dB]: 27.29 -> 28.63 | Improved |
| use sophisticated frame interpolation (DAIN) instead of linear interpolation | Deep internal TSR network · Challenging videos | PSNR [dB]: 26.99 -> 27.09 | No clear effect |
Closest papers by meaning
- Robust Video Super-Resolution With Learned Temporal Dynamics (ICCV 2017 · similarity 0.90)
- Temporal Modulation Network for Controllable Space-Time Video Super-Resolution (CVPR 2021 · similarity 0.90)
- Building an End-to-End Spatial-Temporal Convolutional Network for Video Super-Resolution (AAAI 2017 · similarity 0.90)
- Learning Spatial-Temporal Implicit Neural Representations for Event-Guided Video Super-Resolution (CVPR 2023 · similarity 0.90)
- Real-Time Video Super-Resolution With Spatio-Temporal Networks and Motion Compensation (CVPR 2017 · similarity 0.90)
- 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.
