Research map / Video / Video segmentation and tracking
Low Light Video Enhancement using Synthetic Data Produced with an Intermediate Domain Mapping
Danai Triantafyllidou, Sean Moran, Steven McDonagh, Sarah Parisot, Gregory Slabaugh
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 supervised term L_sup using the intermediate domain B-C pairs (ablation) | CycleGAN B-C · DRV test | PSNR (dB): 24.0 -> 27.2 | Improved |
| add real data to synthetic-only training (ablation) | SIDGAN forward model · DRV static test | PSNR (dB): 21.53 -> 28.94 | Improved |
| add synthetic dynamic data to real data | SID Motion · DRV static test | PSNR (dB): 28.26 -> 28.94 | Improved |
| add synthetic dynamic data to real data | SID Motion · DRV static test | SSIM: 0.815 -> 0.830 | Improved |
| add synthetic data to real data | SID Motion · DRV static video | TPSNR (dB): 38.31 -> 39.34 | Improved |
| add synthetic data to real data | SID Motion · DRV static video | TSSIM: 0.974 -> 0.966 | Hurt |
| add synthetic data with 2% real data (ablation) | SID Motion · DRV dynamic video | warping error x10^-5: 55.9 -> 31.2 | Improved |
| add synthetic data with 20% real data (ablation) | SID Motion · DRV dynamic video | warping error x10^-5: 35.6 -> 32.2 | Improved |
| add synthetic data with 100% real data (ablation) | SID Motion · DRV dynamic video | warping error x10^-5: 29.3 -> 28.2 | Improved |
| add synthetic data at 2% real (ablation) | SID Motion · DRV 2% real | PSNR (dB): 17.70 -> 22.32 | Improved |
| add synthetic data at 5% real (ablation) | SID Motion · DRV 5% real | PSNR (dB): 21.35 -> 23.35 | Improved |
| add synthetic data at 10% real (ablation) | SID Motion · DRV 10% real | PSNR (dB): 24.04 -> 25.19 | Improved |
Closest papers by meaning
- Temporally Consistent Enhancement of Low-Light Videos via Spatial-Temporal Compatible Learning (IJCV 2024 · similarity 0.88)
- Enhancing Low-Light Images: A Synthetic Data Perspective on Practical and Generalizable Solutions (AAAI 2025 · similarity 0.86)
- Binarized Low-light Raw Video Enhancement (CVPR 2024 · similarity 0.85)
- Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset (CVPR 2025 · similarity 0.85)
- Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach (CVPR 2024 · similarity 0.85)
- RGB-Only Supervised Camera Parameter Optimization in Dynamic Scenes (NeurIPS 2025 · similarity 0.84)
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.
