Research map / Video / Video segmentation and tracking
Boosting Event Stream Super-Resolution with a Recurrent Neural Network
Wenming Weng, Yueyi Zhang, Zhiwei Xiong
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 coordinate relocation instead of bicubic upsampling (ablation) | RecEvSR · ENFS-syn 4x | RMSE: 0.655 -> 0.653 | Improved |
| use coordinate relocation instead of bicubic upsampling (ablation) | RecEvSR · ENFS-syn | upsampling time (ms): 217.3 -> 0.2 | Improved |
| add coordinate relocation to EventZoom (ablation) | EventZoom · ENFS-syn 8x | RMSE: 2.385 -> 0.823 | Improved |
| replace EventZoom refinement with RecEvSR (cr upsampling) (ablation) | EventZoom-cr · ENFS-syn 4x | RMSE: 0.833 -> 0.653 | Improved |
| add local temporal correlation (LTC) (ablation) | RecEvSR · ENFS-syn 2x | RMSE: 0.704 -> 0.695 | Improved |
| add global temporal correlation (GTC) (ablation) | RecEvSR · ENFS-syn 2x | RMSE: 0.697 -> 0.695 | Improved |
| add gated temporal fusion (GTF) (ablation) | RecEvSR · ENFS-syn 2x | RMSE: 0.701 -> 0.695 | Improved |
| add adaptive spatiotemporal fusion (ASTF) (ablation) | RecEvSR · ENFS-syn 2x | RMSE: 0.698 -> 0.695 | Improved |
| lengthen training sequence from 3 to 9 (ablation) | RecEvSR · ENFS-syn 2x | RMSE: 0.699 -> 0.689 | Improved |
| lengthen training sequence from 9 to 12 (ablation) | RecEvSR · ENFS-syn 2x | RMSE: 0.689 -> 0.687 | Improved |
| add flip data augmentation (ablation) | RecEvSR · ENFS-syn 2x | RMSE: 0.693 -> 0.688 | Improved |
| RecEvSR instead of EventZoom under added noise | RecEvSR · ENFS-syn 4x (30% noise) | RMSE: 1.093 -> 0.676 | Improved |
Closest papers by meaning
- Event Stream Super-Resolution via Spatiotemporal Constraint Learning (ICCV 2021 · similarity 0.92)
- Learning Spatial-Temporal Implicit Neural Representations for Event-Guided Video Super-Resolution (CVPR 2023 · similarity 0.90)
- Bilateral Event Mining and Complementary for Event Stream Super-Resolution (CVPR 2024 · similarity 0.89)
- EvEnhancer: Empowering Effectiveness, Efficiency and Generalizability for Continuous Space-Time Video Super-Resolution with Events (CVPR 2025 · similarity 0.87)
- Event-based Video Super-Resolution via State Space Models (CVPR 2025 · similarity 0.87)
- Leveraging Asynchronous Spiking Neural Networks for Ultra Efficient Event-Based Visual Processing (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.
