Hierarchical Fusion of Local and Global Visual Features with Mixture-of-Experts for Remote Sensing Image Scene Classification
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915727485698048 |
|---|---|
| author | Tang, Yuanhao Zou, Xuechao Hu, Zhengpei Xing, Junliang Zhang, Chengkun Huang, Jianqiang |
| author_facet | Tang, Yuanhao Zou, Xuechao Hu, Zhengpei Xing, Junliang Zhang, Chengkun Huang, Jianqiang |
| contents | Remote sensing image scene classification remains a challenging task, primarily due to the complex spatial structures and multi-scale characteristics of ground objects. Although CNN-based methods excel at extracting local inductive biases, and Mamba-based approaches demonstrate impressive capabilities in efficiently capturing global sequential context, relying on a single paradigm restricts the model's ability to simultaneously characterize fine-grained textures and complex spatial structures. To tackle this, we propose a parallel heterogeneous encoder, a hierarchical fusion module designed to achieve effective local-global co-representation. It consists of two parallel pathways: a local visual encoder for extracting multi-scale local visual features, and a global visual encoder for capturing efficient global visual features. The core innovation lies in its hierarchical fusion module, which progressively aggregates multi-scale features from both pathways, enabling dynamic cross-level feature interaction and contextual reconstruction to produce highly discriminative representations. These fused features are then adaptively routed through a mixture-of-experts classifier head, which dynamically dispatches them to the most suitable experts for fine-grained scene recognition. Experiments on AID, NWPU-RESISC45, and UC Merced show that our model achieves 93.72%, 95.54%, and 96.92% accuracy, surpassing SOTA methods with an optimal balance of performance and efficiency. Code is available at https://anonymous.4open.science/r/classification-41DF. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_27155 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Hierarchical Fusion of Local and Global Visual Features with Mixture-of-Experts for Remote Sensing Image Scene Classification Tang, Yuanhao Zou, Xuechao Hu, Zhengpei Xing, Junliang Zhang, Chengkun Huang, Jianqiang Computer Vision and Pattern Recognition Remote sensing image scene classification remains a challenging task, primarily due to the complex spatial structures and multi-scale characteristics of ground objects. Although CNN-based methods excel at extracting local inductive biases, and Mamba-based approaches demonstrate impressive capabilities in efficiently capturing global sequential context, relying on a single paradigm restricts the model's ability to simultaneously characterize fine-grained textures and complex spatial structures. To tackle this, we propose a parallel heterogeneous encoder, a hierarchical fusion module designed to achieve effective local-global co-representation. It consists of two parallel pathways: a local visual encoder for extracting multi-scale local visual features, and a global visual encoder for capturing efficient global visual features. The core innovation lies in its hierarchical fusion module, which progressively aggregates multi-scale features from both pathways, enabling dynamic cross-level feature interaction and contextual reconstruction to produce highly discriminative representations. These fused features are then adaptively routed through a mixture-of-experts classifier head, which dynamically dispatches them to the most suitable experts for fine-grained scene recognition. Experiments on AID, NWPU-RESISC45, and UC Merced show that our model achieves 93.72%, 95.54%, and 96.92% accuracy, surpassing SOTA methods with an optimal balance of performance and efficiency. Code is available at https://anonymous.4open.science/r/classification-41DF. |
| title | Hierarchical Fusion of Local and Global Visual Features with Mixture-of-Experts for Remote Sensing Image Scene Classification |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.27155 |