Hyperspectral image super-resolution (HSI SR) focuses on enhancing the spatial resolution of HSIs while preserving their inherent spectral information. Existing single-image HSI SR methods still suffer from blurred spatial edges and spectral distortion. Although numerous spatial–spectral enhancement networks can enhance spatial–spectral feature extraction, they often lead to a cumbersome network architecture. To address these issues, we propose a Spatial–Spectral Distillation Mamba Network, called SSDM-Net, for HSI SR, which contains a main reconstruction branch and two training-only auxiliary branches for spatial and spectral knowledge distillation. Specifically, the spatial and spectral auxiliary branches, which are utilized exclusively during training, provide edge-aware guidance and capture spectral correlations, respectively. During training, the spatial–spectral knowledge is transferred to the main branch. During inference, the auxiliary branches are removed, improving reconstruction quality without extra computational burden. In the main branch, a Mamba-based spatial–spectral global enhancement module processes spatial and latent inter-channel sequences using selective scanning whose cost is linear in the processed sequence lengths when the feature dimensions are fixed. In addition, a dynamic loss weighting strategy is developed to balance reconstruction, distillation, and auxiliary losses during optimization. Comprehensive experiments conducted on the CAVE and Houston datasets with three scale factors demonstrate that SSDM-Net produces more accurate reconstruction results than existing representative HSI SR methods. Cross-dataset experiments on the Harvard dataset further suggest that the method can maintain competitive reconstruction performance under the evaluated cross-dataset settings.

Infrared and visible image fusion combines complementary thermal and structural information from the two modalities into a single composite image. Existing methods have two critical limitations: (1) inadequate utilization of visible structural information causes blurred edges, and (2) modality-specific and shared responses are not always separately represented when learning the fusion weights. To address these problems, an edge-aware dynamic convolution and cross-modal relation-aware fusion network is proposed to fully exploit visible structural details and explicitly model complementary information. Specifically, an edge-aware dynamic convolution is designed to extract edge features from visible images for generating dynamic convolution kernels, which adaptively enhance edge details during fusion. Then, a cross-modal relation-aware adaptive fusion module is designed to explicitly compute difference and interaction features of infrared and visible features and predicts adaptive fusion weights to balance modality-specific information. In addition, a complementarity consistency loss is introduced to jointly constrain intensity, edge and complementary information. Experiments on several public datasets demonstrate that the proposed method provides better visual quality and quantitative performance.