MobileFusion: Mobile-Friendly Infrared and Visible Image Fusion via Structural Re-parameterization

Yufa Duan, Jialing Huang, Yingying Wang, Weimin Cai, Xinghao Ding, Xiaotong Tu
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:26949-26964, 2026.

Abstract

Deep neural networks have recently advanced infrared and visible image fusion (IVIF), but most existing methods rely on sophisticated yet redundant designs, which hinder real-time deployment on mobile devices with limited compute and memory. In this paper, we present MobileFusion, an extremely lightweight and effective convolutional framework that achieves high-quality fusion under strict resource constraints. MobileFusion leverages a novel re-parameterizable multi-branch convolution module to promote cross-modal interactions during training while collapsing into a single-path operator for fast inference. It further incorporates a lightweight attention module to enhance context awareness, together with a re-parameterized feed-forward network to improve feature expressiveness. Extensive experiments demonstrate that MobileFusion delivers a favorable trade-off between fusion quality and computational efficiency, enabling real-time and high-quality IVIF on resource-constrained platforms. The source code is available at https://github.com/sucessfullys/MobileFusion.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-duan26a, title = {{M}obile{F}usion: Mobile-Friendly Infrared and Visible Image Fusion via Structural Re-parameterization}, author = {Duan, Yufa and Huang, Jialing and Wang, Yingying and Cai, Weimin and Ding, Xinghao and Tu, Xiaotong}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {26949--26964}, 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/duan26a/duan26a.pdf}, url = {https://proceedings.mlr.press/v306/duan26a.html}, abstract = {Deep neural networks have recently advanced infrared and visible image fusion (IVIF), but most existing methods rely on sophisticated yet redundant designs, which hinder real-time deployment on mobile devices with limited compute and memory. In this paper, we present MobileFusion, an extremely lightweight and effective convolutional framework that achieves high-quality fusion under strict resource constraints. MobileFusion leverages a novel re-parameterizable multi-branch convolution module to promote cross-modal interactions during training while collapsing into a single-path operator for fast inference. It further incorporates a lightweight attention module to enhance context awareness, together with a re-parameterized feed-forward network to improve feature expressiveness. Extensive experiments demonstrate that MobileFusion delivers a favorable trade-off between fusion quality and computational efficiency, enabling real-time and high-quality IVIF on resource-constrained platforms. The source code is available at https://github.com/sucessfullys/MobileFusion.} }
Endnote
%0 Conference Paper %T MobileFusion: Mobile-Friendly Infrared and Visible Image Fusion via Structural Re-parameterization %A Yufa Duan %A Jialing Huang %A Yingying Wang %A Weimin Cai %A Xinghao Ding %A Xiaotong Tu %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-duan26a %I PMLR %P 26949--26964 %U https://proceedings.mlr.press/v306/duan26a.html %V 306 %X Deep neural networks have recently advanced infrared and visible image fusion (IVIF), but most existing methods rely on sophisticated yet redundant designs, which hinder real-time deployment on mobile devices with limited compute and memory. In this paper, we present MobileFusion, an extremely lightweight and effective convolutional framework that achieves high-quality fusion under strict resource constraints. MobileFusion leverages a novel re-parameterizable multi-branch convolution module to promote cross-modal interactions during training while collapsing into a single-path operator for fast inference. It further incorporates a lightweight attention module to enhance context awareness, together with a re-parameterized feed-forward network to improve feature expressiveness. Extensive experiments demonstrate that MobileFusion delivers a favorable trade-off between fusion quality and computational efficiency, enabling real-time and high-quality IVIF on resource-constrained platforms. The source code is available at https://github.com/sucessfullys/MobileFusion.
APA
Duan, Y., Huang, J., Wang, Y., Cai, W., Ding, X. & Tu, X.. (2026). MobileFusion: Mobile-Friendly Infrared and Visible Image Fusion via Structural Re-parameterization. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:26949-26964 Available from https://proceedings.mlr.press/v306/duan26a.html.

Related Material