Research map / Image generation / Image restoration
Ultra-High-Definition Low-Light Image Enhancement: A Benchmark and Transformer-Based Method
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 LLFormer instead of Restormer | LLIE · UHD-LOL4K | PSNR: 36.9094 -> 37.3340 | Helped |
| Use LLFormer instead of Restormer | LLIE · UHD-LOL8K | PSNR: 35.0568 -> 35.4313 | Helped |
| Use LLFormer instead of Restormer | LLIE · LOL | PSNR: 22.3652 -> 23.6491 | Helped |
| Use LLFormer instead of Restormer | LLIE · MIT-Adobe FiveK | PSNR: 24.9228 -> 25.7528 | Helped |
| Add axis-based attention and DGFN instead of Resblock (ablation) | LLFormer transformer block · UHD-LOL4K | PSNR: 31.92 -> 36.20 | Helped |
| Add depth-wise convolution to attention (ablation) | LLFormer A-MSA · UHD-LOL4K | PSNR: 35.31 -> 36.20; 0.89 | Helped |
| Add depth-wise convolution to DGFN (ablation) | LLFormer DGFN · UHD-LOL4K | PSNR: 35.83 -> 36.20; 0.75 | Helped |
| Add the dual-gated mechanism (ablation) | LLFormer DGFN · UHD-LOL4K | PSNR: 32.78 -> 36.20; 3.42 | Helped |
| Replace conventional FFN with DGFN (ablation) | LLFormer transformer block · UHD-LOL4K | PSNR: 23.33 -> 36.20 | Helped |
| Add weighted skip connections (ablation) | LLFormer · UHD-LOL4K | PSNR: 35.45 -> 36.20 | Helped |
| Add the 1×1 skip-fusion convolution (ablation) | LLFormer · UHD-LOL4K | PSNR: 35.90 -> 36.20 | Helped |
| Add CAFB at both the head and tail (ablation) | LLFormer · UHD-LOL4K | PSNR: 35.10 -> 36.20 | Helped |
Closest papers by meaning
- Enhancing Low-Light Images: A Synthetic Data Perspective on Practical and Generalizable Solutions (AAAI 2025 · similarity 0.88)
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 2025 · similarity 0.87)
- DPLUT: Unsupervised Low-light Image Enhancement with Lookup Tables and Diffusion Priors (AAAI 2025 · similarity 0.87)
- Luminance-Aware Statistical Quantization: Unsupervised Hierarchical Learning for Illumination Enhancement (NeurIPS 2025 · similarity 0.86)
- Interpretable Unsupervised Joint Denoising and Enhancement for Real-World low-light Scenarios (ICLR 2025 · similarity 0.86)
- Improving Visual and Downstream Performance of Low-Light Enhancer with Vision Foundation Models Collaboration (CVPR 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.