AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

ReCoFuse: Ultra-Robust Image Fusion via Restorative Multi-Modal Diffusion Reciprocal Coupling

Hao Zhang, Shuhan Yang, Linfeng Tang, Xunpeng Yi, Jiayi Ma

CVPR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multi-modality image fusion

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 changedWhereMeasuredEffect
add TIM cross-modal coupling to the diffusion restoration branches (ablation)ReCoFuse visible restoration · MFNetPSNR: 24.512 -> 25.205Improved
add TIM cross-modal coupling to the diffusion restoration branches (ablation)ReCoFuse infrared restoration · MFNetFID: 39.190 -> 25.074Improved
insert TIM at each sampling step instead of late fusion after sampling (ablation)ReCoFuse · MFNet fusionMI: MI 3.083 -> 3.114Improved
add time embedding to TIM (ablation)ReCoFuse · MFNet fusionMI: 2.464 -> 3.114Improved
alternating regularization instead of separate-manner optimization (ablation)ReCoFuse · MFNet fusionMI: 2.553 -> 3.114Improved
alternating regularization instead of joint optimization (ablation)ReCoFuse · MFNet fusionMI: 2.972 -> 3.114Improved
alternating regularization instead of alternately freezing one module (decoupling) (ablation)ReCoFuse · MFNet fusionMI: 2.000 -> 3.114Improved
use contrast loss instead of proportional intensity loss (ablation)ReCoFuse · MFNet fusionMI: 2.436 -> 3.114Improved
use texture loss instead of proportional gradient loss (ablation)ReCoFuse · MFNet fusionMI: 2.458 -> 3.114Improved
use ReCoFuse fused images for detectionYOLO-v5 on LLVIP · LLVIP detectionmAP@.75: 0.715 -> 0.718 (IR only vs ReCoFuse)Improved
use ReCoFuse fused images for segmentationSegNeXt on MFNet · MFNet segmentationmIoU: 54.94 -> 57.67 (VIS only vs ReCoFuse)Improved
ReCoFuse versus baselines on real-world AWMM-100kReCoFuse · AWMM-100kMI: 3.095 vs Diff-IF 3.826Mixed

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy