Research map / Video / Video segmentation and tracking
Building an End-to-End Spatial-Temporal Convolutional Network for Video Super-Resolution
Read the paperCheck 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 multi-scale MC-LSTM instead of single-scale C-LSTM (ablation) | STCN · Hollywood2 x5 | PSNR (dB): 32.38 -> 32.56 | Helped |
| use multi-scale MC-LSTM instead of single-scale C-LSTM (ablation) | STCN · Hollywood2 x2 | PSNR (dB): 41.27 -> 41.51 | Helped |
| use bidirectional temporal component instead of unidirectional (ablation) | STCN · Hollywood2 x5 | PSNR (dB): 32.43 -> 32.56 | Helped |
| use bidirectional temporal component instead of unidirectional (ablation) | STCN · Hollywood2 x2 | PSNR (dB): 41.39 -> 41.51 | Helped |
| increase temporal radius T beyond 3 (ablation) | STCN · Hollywood2 | PSNR (dB) | No clear effect |
| STCN with spatio-temporal modeling vs DEL multi-frame | STCN · TriVideos city x4 | PSNR (dB): 24.89 -> 26.01 | Helped |
| STCN vs single-frame VDSR | STCN · Hollywood2 x4 | PSNR (dB): 34.20 -> 34.58 | Helped |
| multi-frame DEL vs single-frame VDSR | video SR · TriVideos penguin x4 | PSNR (dB): 32.59 -> 31.87 | Hurt |
Closest papers by meaning
- Real-Time Video Super-Resolution With Spatio-Temporal Networks and Motion Compensation (CVPR 2017 · similarity 0.91)
- Spatial-Temporal Space Hand-in-Hand: Spatial-Temporal Video Super-Resolution via Cycle-Projected Mutual Learning (CVPR 2022 · similarity 0.91)
- Residual Invertible Spatio-Temporal Network for Video Super-Resolution (AAAI 2019 · similarity 0.90)
- Recurrent Back-Projection Network for Video Super-Resolution (CVPR 2019 · similarity 0.90)
- Frame-Recurrent Video Super-Resolution (CVPR 2018 · similarity 0.89)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · 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.