Research map / Video / Video segmentation and tracking
NIR-assisted Video Enhancement via Unpaired 24-hour Data
Muyao Niu, Zhihang Zhong, Yinqiang Zheng
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 the physics-inspired relighting algorithm for pseudo nighttime NIR (ablation) | NVEU · FMSVD | PI: 5.052 -> 4.786 | Improved |
| add noise GAN to pseudo nighttime VIS (ablation) | NVEU · FMSVD | NIQE: 8.837 -> 5.469 | Improved |
| add resampling trick for day-to-night brightness (ablation) | NVEU · FMSVD | PI: 5.679 -> 4.786 | Improved |
| add saturation loss (ablation) | NVEU · FMSVD | PI: 5.111 -> 4.786 | Improved |
| use relighting instead of CycleGAN to produce pseudo nighttime NIR (ablation) | NVEU · FMSVD | PI: 5.146 -> 4.786 | Improved |
| add relighting on third-party static dataset (ablation) | NVEU · Third-party dataset | PI: 3.039 -> 2.819 | Improved |
| add saturation loss on third-party dataset (ablation) | NVEU · Third-party dataset | PI: 2.823 -> 2.819 | No clear effect |
| retrain DRF and DVN with data generated by the stage-one pipeline | DVN · FMSVD | PI: 8.282 -> 5.214 | Improved |
| use NVEU network instead of retrained DRF or DVN on the same pseudo data | NVEU · FMSVD | PI: 5.195 -> 4.786 | Improved |
Closest papers by meaning
- An Integrated Enhancement Solution for 24-Hour Colorful Imaging (AAAI 2020 · similarity 0.88)
- Dancing in the Dark: A Benchmark towards General Low-light Video Enhancement (ICCV 2023 · similarity 0.87)
- Unrolled Decomposed Unpaired Learning for Controllable Low-Light Video Enhancement (ECCV 2024 · similarity 0.87)
- NightRain: Nighttime Video Deraining via Adaptive-Rain-Removal and Adaptive-Correction (AAAI 2024 · similarity 0.86)
- Exploring the Common Appearance-Boundary Adaptation for Nighttime Optical Flow (ICLR 2024 · similarity 0.85)
- VQCNIR: Clearer Night Image Restoration with Vector-Quantized Codebook (AAAI 2024 · 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.
