Research map / Image generation / Image restoration
Semantically Contrastive Learning for Low-Light Image Enhancement
Read the paperCheck 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 |
|---|---|---|---|
| Use SCL-LLE instead of EnlightenGAN | Low-light image enhancement · DICM, LIME, MEF, VV, NPE average | NIQE: 3.47 -> 3.46 | Helped |
| Use SCL-LLE instead of Zero-DCE | Low-light image enhancement · DICM, LIME, MEF, VV, NPE average | UNIQUE: 0.81 -> 0.85 | Helped |
| Use SCL-LLE instead of Zero-DCE | Low-light image enhancement · DICM, LIME, MEF, VV, NPE average | User Study: 2.70 -> 2.46 | Helped |
| Use SCL-LLE instead of EnlightenGAN | Low-light image enhancement · SICE Part2 | PSNR: 17.82 -> 17.95 | Helped |
| Use SCL-LLE instead of RetinexNet | Low-light image enhancement · SICE Part2 | SSIM: 0.66 -> 0.68 | Helped |
| Add contrastive learning loss (ablation) | SCL-LLE · DICM, LIME, MEF, VV, NPE average | NIQE: 4.27 -> 3.46 | Helped |
| Add semantic brightness consistency loss (ablation) | SCL-LLE · DICM, LIME, MEF, VV, NPE average | UNIQUE: 0.82 -> 0.85 | Helped |
| Add feature retention loss (ablation) | SCL-LLE · DICM, LIME, MEF, VV, NPE average | UNIQUE: 0.80 -> 0.85 | Helped |
| Add negative samples (ablation) | SCL-LLE · DICM, LIME, MEF, VV, NPE average | UNIQUE: 0.78 -> 0.85 | Helped |
| Add overexposed negative samples (ablation) | SCL-LLE · DICM, LIME, MEF, VV, NPE average | UNIQUE: 0.74 -> 0.85 | Helped |
| Add underexposed negative samples (ablation) | SCL-LLE · DICM, LIME, MEF, VV, NPE average | NIQE: 4.38 -> 3.46 | Helped |
Closest papers by meaning
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 2025 · similarity 0.88)
- Improving Visual and Downstream Performance of Low-Light Enhancer with Vision Foundation Models Collaboration (CVPR 2025 · similarity 0.88)
- Luminance-Aware Statistical Quantization: Unsupervised Hierarchical Learning for Illumination Enhancement (NeurIPS 2025 · similarity 0.87)
- Dispel Darkness for Better Fusion: A Controllable Visual Enhancer based on Cross-modal Conditional Adversarial Learning (CVPR 2024 · similarity 0.87)
- Interpretable Unsupervised Joint Denoising and Enhancement for Real-World low-light Scenarios (ICLR 2025 · similarity 0.86)
- Enhancing Low-Light Images: A Synthetic Data Perspective on Practical and Generalizable Solutions (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.