TCTFusion: A Triple-Branch Cross-Modal Transformer for Adaptive Infrared and Visible Image Fusion

Electronics, Год журнала: 2025, Номер 14(4), С. 731 - 731

Опубликована: Фев. 13, 2025

Infrared-visible image fusion (IVIF) is an important part of multimodal (MMF). Our goal to combine useful information from infrared and visible sources produce strong, detailed, fused images that help people understand scenes better. However, most existing methods based on convolutional neural networks extract cross-modal local features without fully utilizing long-range contextual information. This limitation reduces performance, especially in complex scenarios. To address this issue, we propose TCTFusion, a three-branch transformer for visible–infrared fusion. The model includes shallow feature module (SFM), frequency decomposition (FDM), aggregation (IAM). three branches specifically receive input infrared, visible, concatenated images. SFM extracts using residual connections with shared weights. FDM then captures low-frequency global across modalities high-frequency within each modality. IAM aggregates complementary features, enabling the full interaction between different modalities. Finally, decoder generates image. Additionally, introduce pixel loss structural significantly improve model’s overall performance. Extensive experiments mainstream datasets demonstrate TCTFusion outperforms other state-of-the-art both qualitative quantitative evaluations.

Язык: Английский

---

High-Throughput Prediction of Metal-Embedded Complex Properties with a New GNN-Based Metal Attention Framework

Journal of Chemical Information and Modeling, Год журнала: 2025, Номер unknown

Опубликована: Фев. 14, 2025

Metal-embedded complexes (MECs), including transition metal (TMCs) and metal-organic frameworks (MOFs), are important in catalysis, materials science, molecular devices due to their unique atom centrality complex coordination environments. However, modeling predicting properties accurately is challenging. A new attention (MA) framework for graph neural networks (GNNs) was proposed address the limitations of traditional methods, which fail differentiate core structures from ordinary covalent bonds. This MA converts heterogeneous graphs into homogeneous ones with distinct features by highlighting key metal-feature through hierarchical pooling a cross-attention. To assess its performance, 11 widely used GNN algorithms, three heterogeneous, were compared. Experimental results indicate significant improvements accuracy: an average 32.07% TMC up 23.01% MOF CO2 absorption. Moreover, tests on framework's robustness regarding data set size variation comparison larger non-MA model show that enhanced performance stems architecture, not merely increasing capacity. The potential offers potent statistical tool optimizing designing like catalysts gas storage systems.

Язык: Английский

---

Multi-modal image fusion of visible and infrared for precise positioning of UAVs in agricultural fields

Computers and Electronics in Agriculture, Год журнала: 2025, Номер 232, С. 110024 - 110024

Опубликована: Фев. 16, 2025

Язык: Английский

---

Rethinking the approach to lightweight multi-branch heterogeneous image fusion frameworks: Infrared and visible image fusion via the parallel Mamba-KAN framework

Optics & Laser Technology, Год журнала: 2025, Номер 185, С. 112612 - 112612

Опубликована: Фев. 17, 2025

Язык: Английский

---

Efficient feature difference-based infrared and visible image fusion for low-light environments

The Visual Computer, Год журнала: 2025, Номер unknown

Опубликована: Фев. 19, 2025

Язык: Английский

---