Research map / Video / Video segmentation and tracking
VSRM: A Robust Mamba-Based Framework for Video Super-Resolution
Dinh Phu Tran, Dao Duy Hung, Daeyoung Kim
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 Mamba modules instead of 3D DW-Conv (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.84 -> 31.09 | Improved |
| use Mamba modules instead of window attention (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.97 -> 31.09 | Improved |
| use Mamba modules instead of full attention (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 31.06 -> 31.09 | No clear effect |
| add Deformable Cross-mamba Alignment (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.87 -> 31.09 | Improved |
| use DCA instead of FGDA (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.92 -> 31.09 | Improved |
| use DCA instead of IA (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 31.00 -> 31.09 | Improved |
| add T2SMB (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.95 -> 31.09 | Improved |
| use forward-only scan in T2SMB instead of forward and backward (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 31.02 -> 31.09 | Improved |
| use TGFN instead of FFN (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.90 -> 31.09 | Improved |
| add 3D DWConv to TGFN (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.81 -> 31.09 | Improved |
| add channel split to TGFN (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 31.05 -> 31.09 | Improved |
| add Frequency Charbonnier-like loss (ablation) | VSRM · REDS4 (ablation setup) | PSNR: 30.97 -> 31.09 | Improved |
Later papers that compared with this method
1 later accepted paper reports VSRM* as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Trajectory-aware Shifted State Space Models for Online Video Super-Resolution (ICLR 2026 · better on 1 of 1 shared result)
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.
- BasicVSR++ — BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond (CVPR 2021) · better on 3 of 3 shared results
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
- Event-based Video Super-Resolution via State Space Models (CVPR 2025 · similarity 0.90)
- Trajectory-aware Shifted State Space Models for Online Video Super-Resolution (ICLR 2026 · similarity 0.89)
- Self-supervised ControlNet with Spatio-Temporal Mamba for Real-world Video Super-resolution (CVPR 2025 · similarity 0.88)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (AAAI 2025 · similarity 0.88)
- SAVSR: Arbitrary-Scale Video Super-Resolution via a Learned Scale-Adaptive Network (AAAI 2024 · similarity 0.87)
- MobileMamba: Lightweight Multi-Receptive Visual Mamba Network (CVPR 2025 · 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.
