Research map / Video / Video segmentation and tracking
Learning Temporal Consistency for Low Light Video Enhancement From Single Images
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 optical-flow-warped image pairs and an output self-consistency loss | SID-style image-based low-light enhancement network · DAVIS synthetic clean | Warping Error (×1e-3): 1.182 -> 1.061 | Helped |
| Add optical-flow-warped image pairs and an output self-consistency loss | SID-style image-based low-light enhancement network · DAVIS synthetic noisy | PSNR: 22.93 -> 24.01 | Helped |
| Add optical-flow-warped image pairs and an output self-consistency loss | SID-style image-based low-light enhancement network · DAVIS synthetic noisy | Warping Error (×1e-3): 1.303 -> 1.024 | Helped |
| Set consistency-loss weight λ to 15 rather than λ = 0.01 (ablation) | Proposed enhancement network · DAVIS synthetic noisy | PSNR: 22.26 -> 24.05 | Helped |
| Increase consistency-loss weight λ from 0.01 to 20 (ablation) | Proposed enhancement network · DAVIS synthetic noisy | Warping Error (×1e-3): 1.356 -> 1.147 | Helped |
| Use λ = 20 instead of λ = 15 (ablation) | Proposed enhancement network · DAVIS synthetic noisy | PSNR: 24.05 -> 24.04 | No clear effect |
| Train with Gaussian noise only rather than Gaussian and Poisson noise (ablation) | Proposed enhancement network · Synthetic noisy data | PSNR: 23.27; 23.29 | No clear effect |
| Train with Poisson noise only rather than Gaussian and Poisson noise (ablation) | Proposed enhancement network · Synthetic noisy data | PSNR: 23.27; 24.04 | No clear effect |
Closest papers by meaning
- Binarized Low-light Raw Video Enhancement (CVPR 2024 · similarity 0.87)
- Zero-Reference Low-Light Enhancement via Physical Quadruple Priors (CVPR 2024 · similarity 0.86)
- Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach (CVPR 2024 · similarity 0.86)
- Seeing Dark Videos via Self-Learned Bottleneck Neural Representation (AAAI 2024 · similarity 0.86)
- Lux Post Facto: Learning Portrait Performance Relighting with Conditional Video Diffusion and a Hybrid Dataset (CVPR 2025 · similarity 0.85)
- Luminance-Aware Statistical Quantization: Unsupervised Hierarchical Learning for Illumination Enhancement (NeurIPS 2025 · 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.