Research map / Probabilistic and scientific modeling / Causal discovery
CWNet: Causal Wavelet Network for Low-Light Image Enhancement
Tongshun Zhang, Pingping Liu, Yubing Lu, Mengen Cai, Zijian Zhang, Zhe Zhang, Qiuzhan Zhou
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 causal inference mechanism (ablation) | CWNet · LSRW-Huawei | PSNR: 20.87 -> 21.53 | Improved |
| add feature extraction (FE) modules (ablation) | CWNet · LSRW-Huawei | PSNR: 20.98 -> 21.53 | Improved |
| add HFEB (ablation) | CWNet · LSRW-Huawei | PSNR: 20.58 -> 21.53 | Improved |
| add LFEB (ablation) | CWNet · LSRW-Huawei | PSNR: 20.41 -> 21.53 | Improved |
| use WTConv instead of standard convolution (ablation) | CWNet · LSRW-Huawei | PSNR: 21.42 -> 21.53 | Improved |
| use FFTConv instead of standard convolution (ablation) | CWNet · LSRW-Huawei | PSNR: 21.36 -> 21.53 | Improved |
| use HF-Mamba instead of VMamba 2D-SSM (ablation) | CWNet · LSRW-Huawei | PSNR: 21.20 -> 21.53 | Improved |
| use semantic maps instead of global features (ablation) | CWNet · LSRW-Huawei | PSNR: 21.48 -> 21.53 | Improved |
| raise semantic loss weight from 0.01 to 0.05 (ablation) | CWNet · LSRW-Huawei | PSNR: 21.53 -> 20.89 | Hurt |
| lower causal loss weight from 0.01 to 0.001 (ablation) | CWNet · LSRW-Huawei | PSNR: 21.53 -> 21.17 | Hurt |
| use 1 negative sample each instead of 2 (ablation) | CWNet · LSRW-Huawei | PSNR: 21.50 -> 21.06 | Hurt |
| raise negative samples from 2 to 3 each (ablation) | CWNet · LSRW-Huawei | PSNR: 21.50 -> 21.34 | Mixed |
Closest papers by meaning
- Semantically Contrastive Learning for Low-Light Image Enhancement (AAAI 2022 · similarity 0.86)
- Temporally Consistent Enhancement of Low-Light Videos via Spatial-Temporal Compatible Learning (IJCV 2024 · similarity 0.85)
- Interpretable Unsupervised Joint Denoising and Enhancement for Real-World low-light Scenarios (ICLR 2025 · similarity 0.85)
- Empowering Low-Light Image Enhancer through Customized Learnable Priors (ICCV 2023 · similarity 0.85)
- Learning Semantic-Aware Knowledge Guidance for Low-Light Image Enhancement (CVPR 2023 · similarity 0.85)
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 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.
