Research map / Vision-language and multimodal / Cross-modal learning
CtrlFuse: Mask-Prompt Guided Controllable Infrared and Visible Image Fusion
Yiming Sun, Yuan Ruan, Qinghua Hu, Pengfei Zhu
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 prompt mask guidance (ablation) | CtrlFuse · MSRS | Qabf: Qabf 0.637 -> 0.685; SSIM 0.933 -> 0.969; PSNR 64.958 -> 64.75 | Mixed |
| add SAM segmentation branch and segmentation loss (ablation) | CtrlFuse · MSRS | Qabf: Qabf 0.671 -> 0.685; SSIM 0.939 -> 0.969 | Improved |
| add visible-branch prompt encoder and fusion module (ablation) | CtrlFuse · MSRS | SSIM: SSIM 0.915 -> 0.969; PSNR 65.064 -> 64.75 | Mixed |
| add infrared-branch prompt encoder and fusion module (ablation) | CtrlFuse · MSRS | SCD: SCD 1.622 -> 1.726; PSNR 65.027 -> 64.75 | Mixed |
| use source feature as support and fused reference as query (ablation) | reference prompt encoder · MSRS | SSIM: SSIM 0.924 -> 0.969; SCD 1.659 -> 1.726 | Improved |
| fine-tune SAM through CtrlFuse prompts | SAM · FMB segmentation | qualitative segmentation | Improved |
| use incomplete or coarse prompt masks | CtrlFuse · fusion visual results | qualitative | No clear effect |
| use CtrlFuse fused images for segmentation versus top baselines | segmentation model · MSRS segmentation | mIoU: 0.7955 (PSFusion) -> 0.7963 | Improved |
| use CtrlFuse fused images for detection versus top baselines | YOLOv5 · DroneVehicle | AP@[0.5:0.95] (All): 0.510 (SDCFusion) -> 0.525 | Improved |
Closest papers by meaning
- Task-driven Image Fusion with Learnable Fusion Loss (CVPR 2025 · similarity 0.86)
- Every SAM Drop Counts: Embracing Semantic Priors for Multi-Modality Image Fusion and Beyond (CVPR 2025 · similarity 0.85)
- MRFS: Mutually Reinforcing Image Fusion and Segmentation (CVPR 2024 · similarity 0.84)
- ControlFusion: A Controllable Image Fusion Network with Language-Vision Degradation Prompts (NeurIPS 2025 · similarity 0.84)
- Revisiting Generative Infrared and Visible Image Fusion Based on Human Cognitive Laws (NeurIPS 2025 · similarity 0.84)
- One Model for ALL: Low-Level Task Interaction Is a Key to Task-Agnostic Image Fusion (CVPR 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.
