September 10, 2025Open Access

MMFNet: A Mamba-Based Multimodal Fusion Network for Semantic Segmentation of Remote Sensing

View Full Paper

Ask AI

Bookmark

MMFNet achieves mean IoU scores of 83.50% and 86.06% on benchmark datasets, outperforming existing methods.
The dual-encoder design combines ResNet-18 for local detail and VMamba for global context in segmentation.
A frequency-aware upsampling module enhances boundary and spatial detail recovery in the segmentation process.
Extensive experiments validate MMFNet's efficiency and accuracy in multimodal semantic segmentation applications.

Discussion

No takes yet. Share an insight, caveat, or question.

Proposed MMFNet shows improved mean IoU scores in remote sensing imagery by leveraging multimodal fusion techniques.

Qiu et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark