Research map / Image generation / Image restoration
FusionBooster: A Unified Image Fusion Boosting Paradigm
Chunyang Cheng, Tianyang Xu, Xiaojun Wu, Hui Li, Xi Li, Josef Kittler
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 the FusionBooster second stage to the backbone (vs enhancing the source images first) (ablation) | GTF · LLVIP (IVIF) | VIF: 0.23 -> 0.88 | Improved |
| add the FusionBooster second stage (vs directly enhancing the fusion result) (ablation) | GTF · LLVIP (IVIF) | VIF: 0.18 -> 0.88 | Improved |
| add the information probe (vs feeding source images to ASE) (ablation) | GTF · LLVIP (IVIF) | VIF: 0.60 -> 0.88 | Improved |
| add the information probe (vs feeding enhanced source images to ASE) (ablation) | GTF · LLVIP (IVIF) | VIF: 0.72 -> 0.88 | Improved |
| add the information probe (vs enhanced source images to ASE) (ablation) | U2Fusion · SCIE (MEIF) | VIF/EI: VIF 2.29 -> 2.51; EI 147.93 -> 134.52 | Mixed |
| add the information probe (vs enhanced source images to ASE) (ablation) | CSR · MFI-WHU (MFIF) | VIF: 1.75 -> 1.79 | No clear effect |
| add FusionBooster | DDcGAN · LLVIP (IVIF) | VIF: 0.764 -> 0.986 | Improved |
| add FusionBooster | GTF · LLVIP (IVIF) | VIF: 0.576 -> 0.884 | Improved |
| add FusionBooster | YDTR · LLVIP (IVIF) | VIF: 0.388 -> 0.532 | Improved |
| add FusionBooster to the fusion images used for detector training | DDcGAN · LLVIP pedestrian detection | mAP50:95: 63.7 -> 69.3 (+5.6) | Improved |
| add FusionBooster | U2Fusion · SCIE (MEIF) | VIF: 1.695 -> 2.506 | Improved |
| add FusionBooster | TLER · SCIE (MEIF) | Qabf: 0.736 -> 0.518 | Hurt |
Closest papers by meaning
- A Task-Guided, Implicitly-Searched and Meta-Initialized Deep Model for Image Fusion (TPAMI 2024 · similarity 0.88)
- Task-driven Image Fusion with Learnable Fusion Loss (CVPR 2025 · similarity 0.85)
- ControlFusion: A Controllable Image Fusion Network with Language-Vision Degradation Prompts (NeurIPS 2025 · similarity 0.84)
- Efficient Rectified Flow for Image Fusion (NeurIPS 2025 · similarity 0.84)
- DCEvo: Discriminative Cross-Dimensional Evolutionary Learning for Infrared and Visible Image Fusion (CVPR 2025 · similarity 0.84)
- Test-Time Dynamic Image Fusion (NeurIPS 2024 · similarity 0.83)
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.
