Research map / Architectures and efficiency / Pruning and efficient tuning
Structured Sparsity Learning for Efficient 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 |
|---|---|---|---|
| apply SSL pruning to bidirectional BasicVSR instead of ASSL pruning | BasicVSR · REDS4 (BI, ×4) | PSNR: 30.74 -> 31.06 | Helped |
| apply SSL pruning to unidirectional BasicVSR instead of ASSL pruning | BasicVSR-uni · Vid4 (BD, ×4) | PSNR: 27.02 -> 27.18 | Helped |
| replace aligned residual-block pruning with RSC (ablation) | SSL pruning of BasicVSR · REDS4 (×4) | PSNR: 30.86 -> 31.06 | Helped |
| use pixel-shuffle pruning rather than halving upsampling filters (ablation) | SSL pruning of BasicVSR · REDS4 (×4) | PSNR: 30.82 -> 30.98 | Helped |
| add Temporal Finetuning (ablation) | SSL pruning of BasicVSR · REDS4 (×4) | PSNR: 30.98 -> 31.06; 0.08 dB | Helped |
| select low-L1 filters globally rather than locally (ablation) | SSL pruning of BasicVSR · REDS4 (×4), pruning ratio 0.5 | PSNR: 30.83 -> 31.06 | Helped |
| select high-L1 filters globally instead of random filters (ablation) | SSL pruning of BasicVSR · REDS4 (×4), pruning ratio 0.5 | PSNR: 30.68 -> 28.90 | Hurt |
| select high-L1 filters locally instead of random filters (ablation) | SSL pruning of BasicVSR · REDS4 (×4), pruning ratio 0.5 | PSNR: 30.68 -> 30.38 | Hurt |
Closest papers by meaning
- Aligned Structured Sparsity Learning for Efficient Image Super-Resolution (NeurIPS 2021 · similarity 0.91)
- Learning Efficient Image Super-Resolution Networks via Structure-Regularized Pruning (ICLR 2022 · similarity 0.89)
- Lightweight Image Super-Resolution via Flexible Meta Pruning (ICML 2024 · similarity 0.87)
- DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution (NeurIPS 2025 · similarity 0.86)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.86)
- Týr-the-Pruner: Structural Pruning LLMs via Global Sparsity Distribution Optimization (NeurIPS 2025 · similarity 0.85)
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.