Research map / Image generation / Image restoration
Deep Multi-Model Fusion for Single-Image Dehazing
Zijun Deng, Lei Zhu, Xiaowei Hu, Chi-Wing Fu, Xuemiao Xu, Qing Zhang, Jing Qin, Pheng-Ann Heng
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 |
|---|---|---|---|
| fuse atmospheric scattering and four layer-separation outputs instead of scattering model alone (ablation) | DM2F-Net · TestA-DCPDN | PSNR: 34.36 -> 35.61 | Improved |
| fuse atmospheric scattering and four layer-separation outputs instead of scattering model alone (ablation) | DM2F-Net · SOTS | PSNR: 32.42 -> 34.29 | Improved |
| fuse four layer separation outputs plus scattering instead of J1 only (ablation) | DM2F-Net · TestA-DCPDN | PSNR: 30.57 -> 35.61 | Improved |
| fuse instead of J2 only (ablation) | DM2F-Net · SOTS | PSNR: 30.92 -> 34.29 | Improved |
| fuse instead of J3 only (ablation) | DM2F-Net · TestA-DCPDN | PSNR: 32.30 -> 35.61 | Improved |
| fuse instead of J4 only (ablation) | DM2F-Net · SOTS | PSNR: 28.85 -> 34.29 | Improved |
| add attentional feature integration module (AFIM) (ablation) | DM2F-Net · TestA-DCPDN | PSNR: 34.71 -> 35.61 | Improved |
| add attentional feature integration module (AFIM) (ablation) | DM2F-Net · SOTS | PSNR: 33.93 -> 34.29 | Improved |
| add attentional feature integration module (AFIM) (ablation) | DM2F-Net · TestA-DCPDN | SSIM: 0.9810 -> 0.9829 | Improved |
| DM2F-Net vs best prior network DCPDN | DM2F-Net · SOTS | PSNR: 28.13 -> 34.29 | Improved |
| DM2F-Net vs DCPDN on real outdoor haze | DM2F-Net · O-HAZE | PSNR: 22.777 -> 25.188 | Improved |
| DM2F-Net vs challenge top team BJTU | DM2F-Net · O-HAZE | SSIM: 0.777 -> 0.777 | No clear effect |
Closest papers by meaning
- FFA-Net: Feature Fusion Attention Network for Single Image Dehazing (AAAI 2020 · similarity 0.90)
- Towards Multi-Domain Single Image Dehazing via Test-Time Training (CVPR 2022 · similarity 0.90)
- Tokenize Image Patches: Global Context Fusion for Effective Haze Removal in Large Images (CVPR 2025 · similarity 0.87)
- PromptHaze: Prompting Real-world Dehazing via Depth Anything Model (AAAI 2025 · similarity 0.86)
- Real-world Image Dehazing with Coherence-based Pseudo Labeling and Cooperative Unfolding Network (NeurIPS 2024 · similarity 0.86)
- Learning Hazing to Dehazing: Towards Realistic Haze Generation for Real-World Image Dehazing (CVPR 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.
