Second, we use the weight-sharing strategy to model the correlations of cross-modal samples to narrow down the modality gap. Finally, we apply the prototype to achieve class-incremental learning to prove the ro
In this paper, we propose $extbf{Multimodal Federated Cross Prototype Learning (MFCPL)}$, a novel approach for MFL under severely missing modalities. Our MFCPL leverages the complete prototypes to provide diverse modality knowledge in modality-shared level with the cross-modal regularization and ...