However, the former fails to capture the deeper semantic information of images effectively, while the latter tends to lose critical detail features when registering high-resolution images. In this ...