Research map / Video / Video segmentation and tracking
Arbitrary-Scale Video Super-Resolution with Structural and Textural Priors
Wei Shang, Dongwei Ren, Wanying Zhang, Yuming Fang, Wangmeng Zuo, Kede Ma
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 |
|---|---|---|---|
| add multi-scale structural and textural VGG prior to B-AVSR (ablation) | B-AVSR · REDS val x4 | PSNR: 29.67 -> 31.03 | Improved |
| add flow-refined cross-attention unit (ablation) | ST-AVSR · REDS val x4 | PSNR: 29.31 -> 31.03 | Improved |
| add flow rectification in the cross-attention unit (ablation) | ST-AVSR · REDS val x4 | PSNR: 30.68 -> 31.03 | Improved |
| add optical flow estimation (vs deformable conv only) (ablation) | ST-AVSR · REDS val x4 | PSNR: 30.39 -> 31.03 | Improved |
| use local cross-attention plus global self-attention instead of concatenation (ablation) | ST-AVSR · REDS val x4 | PSNR: 30.36 -> 31.03 | Improved |
| add global self-attention (ablation) | ST-AVSR · REDS val x4 | PSNR: 30.29 -> 31.03 | Improved |
| increase future window L from 0 to 2 (ablation) | ST-AVSR · REDS val x4 | PSNR: 30.12 -> 31.03 | Improved |
| increase future window L from 2 to 3 (ablation) | ST-AVSR · REDS val x4 | PSNR: 31.03 -> 31.05 | No clear effect |
| use ST-AVSR instead of implicit-representation AVSR/AISR methods | ST-AVSR · Vid4 x4 inference time | inference time (s): 0.1242 -> 0.0495 | Improved |
| use ST-AVSR on extreme scale x7.2/6 | ST-AVSR · Vid4 x7.2/6 | PSNR: 21.72 -> 21.60 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MoTIF — MoTIF: Learning Motion Trajectories with Local Implicit Neural Functions for Continuous Space-Time Video Super-Resolution (ICCV 2023) · better on 3 of 3 shared results
- VideoINR — VideoINR: Learning Video Implicit Neural Representation for Continuous Space-Time Super-Resolution (CVPR 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.89)
- Arbitrary-Scale Video Super-resolution Guided by Dynamic Context (AAAI 2024 · similarity 0.88)
- VideoGigaGAN: Towards Detail-rich Video Super-Resolution (CVPR 2025 · similarity 0.88)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (AAAI 2025 · similarity 0.87)
- NSARM: Next-Scale Autoregressive Modeling for Robust Real-World Image Super-Resolution (NeurIPS 2026 · similarity 0.87)
- SkipSR: Faster Super-Resolution with Token Skipping (NeurIPS 2026 · 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.
