AI Paper Atlas

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

Infrared and Visible Image Fusion: From Data Compatibility to Task Adaption

Jinyuan Liu, Guanyao Wu, Zhu Liu, Di Wang, Zhiying Jiang, Long Ma, Wei Zhong, Xin Fan, Risheng Liu

TPAMI 2024 · Vision-language and multimodal › Cross-modal learning · approach: Multi-modality image fusion · 260 citations (OpenAlex)

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
use detection-oriented fusion TarDAL instead of the other fusion methodsfusion for YOLOv5 detection · M3FD detection (YOLOv5 on fused images)mAP@0.5: 0.598 vs 0.587Improved
use detection-oriented TarDAL instead of CoCoNetfusion for detection · M3FD detection (YOLOv5 on fused images)mAP@0.5: 0.469 -> 0.598Improved
use perception-guided fusion (SegMiF, TIMFusion) rather than visual-only DDcGANfusion for detection · M3FD detection (YOLOv5 on fused images)mAP@0.5: 0.585 and 0.584 vs 0.528Improved
optimize with feature-level contrastive loss (CoCoNet)image fusion · TNO and RoadScene fusion (pretrained models)EN, SD: 7.735 EN and 64.44 SDImproved
optimize fusion with segmentation semanticsPromptF · TNO and RoadScene fusion (pretrained models)VIF: 0.700Improved
add ground-truth deformation supervisionSuperFusion · Misaligned fusion RoadSceneMI: 2.25Improved
use style-transfer registration then fusionUMFusion and IMF · Misaligned fusion M3FDVIF, SCD: 0.87 VIF; 1.54 SCDMixed
use latent feature space registration instead of style transferRFVIF and MURF · Misaligned fusion RoadSceneEN, SF, SD, AG: 46.05 SD; 8.42 AGImproved
use dual discriminators instead of a single discriminatorGAN fusion · IVIFmodality balanceImproved

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