Research map / Video / Video segmentation and tracking
Flow-Guided Sparse Transformer for Video Deblurring
Jing Lin, Yuanhao Cai, Xiaowan Hu, Haoqian Wang, Youliang Yan, Xueyi Zou, Henghui Ding, Yulun Zhang, Radu Timofte, Luc Van Gool
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 FGST rather than the strongest reported baseline | Video deblurring · DVD | PSNR: 32.80 -> 33.36 | Improved |
| Use FGST rather than the strongest reported baseline | Video deblurring · GOPRO | PSNR: 32.10 -> 32.90 | Improved |
| Add RE to the baseline (ablation) | FGST · GOPRO | PSNR: 31.18 -> 32.34 (+1.16 dB) | Improved |
| Add FGSW-MSA to the baseline (ablation) | FGST · GOPRO | PSNR: 31.18 -> 32.84 (+1.66 dB) | Improved |
| Add both RE and FGSW-MSA (ablation) | FGST · GOPRO | PSNR: 31.18 -> 32.90 (+1.72 dB) | Improved |
| Replace FGSW-MSA with global MSA (ablation) | FGST · GOPRO | PSNR: 31.18 -> 29.20 (-1.98 dB) | Hurt |
| Replace FGSW-MSA with local W-MSA (ablation) | FGST · GOPRO | PSNR: 31.18 -> 31.71 (+0.53 dB) | Improved |
| Use FGSW-MSA rather than FGS-MSA (ablation) | FGST · GOPRO | PSNR: 32.48 -> 32.84 | Improved |
| Use FGSW-MSA rather than DeConv (ablation) | FGST · GOPRO | PSNR: 32.35 -> 32.84 | Improved |
| Use FGSW-MSA rather than FGDeConv (ablation) | FGST · GOPRO | PSNR: 32.59 -> 32.84 | Improved |
| Replace pre-warping with FGSW-MSA (ablation) | FGST · GOPRO | PSNR: 32.54 -> 32.84 (0.30 dB) | Improved |
| Set FGSW-MSA window size to 3×3 rather than 1×1 (ablation) | FGST · GOPRO | PSNR: 32.48 -> 32.90 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ARVo — ARVo: Learning All-Range Volumetric Correspondence for Video Deblurring (CVPR 2021) · 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
- Blur-aware Spatio-temporal Sparse Transformer for Video Deblurring (CVPR 2024 · similarity 0.91)
- FMA-Net: Flow-Guided Dynamic Filtering and Iterative Feature Refinement with Multi-Attention for Joint Video Super-Resolution and Deblurring (CVPR 2024 · similarity 0.87)
- Bidirectional Sparse Attention for Faster Video Diffusion Training (NeurIPS 2026 · similarity 0.87)
- Sparse Video-Gen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity (ICML 2025 · similarity 0.86)
- Adaptive Window Pruning for Efficient Local Motion Deblurring (ICLR 2024 · similarity 0.86)
- Motion-adaptive Transformer for Event-based Image Deblurring (AAAI 2025 · similarity 0.86)
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.
