Research map / Video / Video segmentation and tracking
Seeing the Unseen: Zooming in the Dark with Event Cameras
Dachun Kai, Zeyu Xiao, Huyue Zhu, Jiaxiao Wang, Yueyi Zhang, Xiaoyan Sun
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 IEE module (illumination-guided event enhancement) (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 28.27 -> 30.28 | Improved |
| add ERE module (event-guided reflectance enhancement) (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 27.31 -> 30.28 | Improved |
| use three IEE scales instead of one (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 28.64 -> 30.28 | Improved |
| use three IEE scales instead of two (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 28.83 -> 30.28 | Improved |
| use multi-scale ERE instead of single-scale (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 28.04 -> 30.28 | Improved |
| add cross-modal fusion in ERE (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 29.78 -> 30.28 | Improved |
| use unsupervised SCI instead of URetinex (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 29.62 -> 30.28 | Improved |
| compute optical flow from reflectance instead of raw frames (ablation) | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 29.85 -> 30.28 | Improved |
| use RetinexEVSR instead of EvTexture | RetinexEVSR · RELED 4x LVSR | FLOPs (G): 1141.1 -> 159.1 | Improved |
| use RetinexEVSR instead of EvTexture | RetinexEVSR · SDSD-in 4x LVSR | PSNR (dB): 27.33 -> 30.28 | Improved |
| use RetinexEVSR vs DP3DF on no-reference NIQE | RetinexEVSR · SDE-in real-world SR (no downsampling) | NIQE: 6.8206 -> 7.0684 | 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.
- EvTexture — EvTexture: Event-driven Texture Enhancement for Video Super-Resolution (ICML 2024) · better on 8 of 8 shared results
- FMA-Net — FMA-Net: Flow-Guided Dynamic Filtering and Iterative Feature Refinement with Multi-Attention for Joint Video Super-Resolution and Deblurring (CVPR 2024) · better on 5 of 5 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
- Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach (CVPR 2024 · similarity 0.89)
- Low-Light Image Enhancement Using Event-Based Illumination Estimation (ICCV 2025 · similarity 0.89)
- Event-guided Unified Framework for Low-light Video Enhancement, Frame Interpolation, and Deblurring (ICCV 2025 · similarity 0.89)
- eRetinexGS: Retinex Modeling for Low-Light Scene Enhancement via Event Streams and 3D Gaussian Splatting (CVPR 2026 · similarity 0.89)
- Event-Guided Consistent Video Enhancement with Modality-Adaptive Diffusion Pipeline (NeurIPS 2025 · similarity 0.88)
- Event-Enhanced Blurry Video Super-Resolution (AAAI 2025 · similarity 0.88)
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.
