Difference-Aware Decision Learning for Multimodal Image Fusion

Hao Pan, Jian Dai, Yuan Sun, Zhenwen Ren, Xingfeng Li
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:95745-95762, 2026.

Abstract

Multimodal image fusion aims to integrate complementary information from different modalities, but cross-modal discrepancies and local conflicts often make modality allocation uncertain, causing information loss or artifact propagation. We address this problem by formulating fusion as an observation-conditioned probabilistic decision-learning problem, where local modality contribution is explicitly modeled as a decision variable. Based on this view, we propose a dIfference-aware Decision-lEArning muLtimodal image fusion paradigm (IDEAL). IDEAL uses cross-modal differences as decision triggers and constructs spatial and spectral decision conditions from multi-scale difference attention, power-spectrum energy, complementary spectra, and spectral-entropy reliability. These conditions are mapped to interpretable contribution policies through a symmetric Beta prior, while uncertainty modulation pulls unreliable decisions toward conservative mixing when evidence is insufficient. Extensive experiments on multiple fusion tasks demonstrate stable and competitive performance against state-of-the-art methods. Code is available at: https://github.com/Pon915/IDEAL-main.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-pan26d, title = {Difference-Aware Decision Learning for Multimodal Image Fusion}, author = {Pan, Hao and Dai, Jian and Sun, Yuan and Ren, Zhenwen and Li, Xingfeng}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {95745--95762}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/pan26d/pan26d.pdf}, url = {https://proceedings.mlr.press/v306/pan26d.html}, abstract = {Multimodal image fusion aims to integrate complementary information from different modalities, but cross-modal discrepancies and local conflicts often make modality allocation uncertain, causing information loss or artifact propagation. We address this problem by formulating fusion as an observation-conditioned probabilistic decision-learning problem, where local modality contribution is explicitly modeled as a decision variable. Based on this view, we propose a dIfference-aware Decision-lEArning muLtimodal image fusion paradigm (IDEAL). IDEAL uses cross-modal differences as decision triggers and constructs spatial and spectral decision conditions from multi-scale difference attention, power-spectrum energy, complementary spectra, and spectral-entropy reliability. These conditions are mapped to interpretable contribution policies through a symmetric Beta prior, while uncertainty modulation pulls unreliable decisions toward conservative mixing when evidence is insufficient. Extensive experiments on multiple fusion tasks demonstrate stable and competitive performance against state-of-the-art methods. Code is available at: https://github.com/Pon915/IDEAL-main.} }
Endnote
%0 Conference Paper %T Difference-Aware Decision Learning for Multimodal Image Fusion %A Hao Pan %A Jian Dai %A Yuan Sun %A Zhenwen Ren %A Xingfeng Li %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-pan26d %I PMLR %P 95745--95762 %U https://proceedings.mlr.press/v306/pan26d.html %V 306 %X Multimodal image fusion aims to integrate complementary information from different modalities, but cross-modal discrepancies and local conflicts often make modality allocation uncertain, causing information loss or artifact propagation. We address this problem by formulating fusion as an observation-conditioned probabilistic decision-learning problem, where local modality contribution is explicitly modeled as a decision variable. Based on this view, we propose a dIfference-aware Decision-lEArning muLtimodal image fusion paradigm (IDEAL). IDEAL uses cross-modal differences as decision triggers and constructs spatial and spectral decision conditions from multi-scale difference attention, power-spectrum energy, complementary spectra, and spectral-entropy reliability. These conditions are mapped to interpretable contribution policies through a symmetric Beta prior, while uncertainty modulation pulls unreliable decisions toward conservative mixing when evidence is insufficient. Extensive experiments on multiple fusion tasks demonstrate stable and competitive performance against state-of-the-art methods. Code is available at: https://github.com/Pon915/IDEAL-main.
APA
Pan, H., Dai, J., Sun, Y., Ren, Z. & Li, X.. (2026). Difference-Aware Decision Learning for Multimodal Image Fusion. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:95745-95762 Available from https://proceedings.mlr.press/v306/pan26d.html.

Related Material