Research map / Image generation / Image restoration
Bilevel Fast Scene Adaptation for Low-Light Image Enhancement
Long Ma, Dian Jin, Nan An, Jinyuan Liu, Xin Fan, Zhongxuan Luo, Risheng Liu
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 adaptive denoising module (ablation) | BL/RBL · LOL | PSNR: 11.1692 -> 15.4120 | Improved |
| add adaptive denoising module (ablation) | BL/RBL · LOL | SSIM: 0.2750 -> 0.4205 | Improved |
| add adaptive denoising module (ablation) | BL/RBL · LOL | LPIPS: 1.1439 -> 0.5444 | Improved |
| add finetuning of decoder on adaptation data (ablation) | BL · VOC | PSNR: 15.7692 -> 16.7280 | Improved |
| add finetuning of decoder on adaptation data (ablation) | BL · VOC | SSIM: 0.6181 -> 0.6325 | Improved |
| add finetuning of decoder on adaptation data (ablation) | BL · VOC | LPIPS: 0.2503 -> 0.1987 | Improved |
| add reinforced meta-initialization of decoder (RBL over BL) | BL · MIT | PSNR: 20.1299 -> 20.6759 | Improved |
| add reinforced meta-initialization of decoder (RBL over BL) | BL · LSRW | SSIM: 0.5933 -> 0.6761 | Improved |
| add reinforced meta-initialization of decoder (RBL over BL) | BL · LSRW | PSNR: 14.8726 -> 15.3892 | Improved |
| add reinforced meta-initialization of decoder (RBL over BL) | BL · VOC | LPIPS: 0.1987 -> 0.2103 | Hurt |
| add reinforced meta-initialization of decoder (RBL over BL) | BL · MIT | NIQE: 4.0318 -> 4.4554 | Hurt |
| add reinforced meta-initialization of decoder (RBL over BL) | BL · LOL | SSIM: 0.7331 -> 0.7728 | Improved |
Closest papers by meaning
- BiProLoRA: Bilevel Prompt LoRA for Real Scene Recovery (CVPR 2026 · similarity 0.88)
- Enhancing Neural Radiance Fields with Adaptive Multi-Exposure Fusion: A Bilevel Optimization Approach for Novel View Synthesis (AAAI 2024 · similarity 0.85)
- Learning With Nested Scene Modeling and Cooperative Architecture Search for Low-Light Vision (TPAMI 2022 · similarity 0.85)
- You Do Not Need Additional Priors or Regularizers in Retinex-Based Low-Light Image Enhancement (CVPR 2023 · similarity 0.85)
- Low-Light Image Enhancement with Normalizing Flow (AAAI 2022 · similarity 0.85)
- Learning a Simple Low-Light Image Enhancer From Paired Low-Light Instances (CVPR 2023 · 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.
