Vision Transformers (ViTs) have shown promise in multimodal fusion image classification, yet face performance challenges in complex remote sensing scenarios. Single fusion frameworks often fail to ...

To overcome this limitation, we propose a novel joint learning method termed the dual-conditionally guided registration and fusion diffusion network (RFDifNet) for unregistered remote sensing images.