Research map / Vision-language and multimodal / Cross-modal learning
P²Fusion: Prompt-based Progressive Infrared-Visible Image Fusion via Dual-Prior Distillation
Yi Shi, Huichao Xie, Yuqing Wang, Mingyu Wang, Kaihui Yang, Yu Liu, Lu Ruitao, Lizhe Li, Junwei Han, Dingwen Zhang
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 visible quality prior (vis prompt) (ablation) | P2Fusion · MSRS/M3FD/FMB/RoadScene fusion | VIF | Improved |
| add infrared saliency prior (ir prompt) (ablation) | P2Fusion · fusion benchmarks | fusion metrics | Improved |
| add both infrared and visible priors (ablation) | P2Fusion · fusion benchmarks | fusion metrics | Improved |
| use GDER instead of simpler structural alternatives (e.g. patch-level loss on P_vis) (ablation) | P2Fusion · fusion benchmarks | fusion metrics | Improved |
| add both Modality and Attention experts in GDER (ablation) | P2Fusion · fusion benchmarks | fusion metrics | Improved |
| use both experts, which have Pearson correlation between them (ablation) | P2Fusion GDER · fusion benchmarks | expert correlation: r=0.46 | Improved |
| use P2Fusion fused images for detection vs previous SOTA | P2Fusion · MSRS detection | mAP50-95: +3.2% | Improved |
| use P2Fusion fused images for detection vs previous SOTA | P2Fusion · M3FD detection | mAP: +0.5% | Improved |
| use P2Fusion on unseen DroneVehicle without training exposure | P2Fusion · DroneVehicle detection (OOD) | mAP50-95: +0.9% | Improved |
| remove visible prompt P_vis from gating (ablation) | P2Fusion GDER · gating visualization | gating distribution | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- LutFuse — LUT-Fuse: Towards Extremely Fast Infrared and Visible Image Fusion via Distillation to Learnable Look-Up Tables (ICCV 2025) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Infrared and Visible Image Fusion: From Data Compatibility to Task Adaption (TPAMI 2024 · similarity 0.90)
- Domain Adaptation Guided Infrared and Visible Image Fusion (AAAI 2026 · similarity 0.89)
- MobileFusion: Mobile-Friendly Infrared and Visible Image Fusion via Structural Re-parameterization (ICML 2026 · similarity 0.89)
- Fusion in Your Way: Aligning Image Fusion with Heterogeneous Demands via Direct Preference Optimization (CVPR 2026 · similarity 0.88)
- Enhancing Infrared Vision: Progressive Prompt Fusion Network and Benchmark (NeurIPS 2025 · similarity 0.88)
- Every SAM Drop Counts: Embracing Semantic Priors for Multi-Modality Image Fusion and Beyond (CVPR 2025 · similarity 0.88)
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.
