Federated Learning for Blind Image Super-Resolution

arXiv:2404.17670v1 Announce Type: cross
Abstract: Traditional blind image SR methods need to model real-world degradations precisely. Consequently, current research struggles with this dilemma by assuming idealized degradations, which leads to limited applicability to actual user data. Moreover, the ideal scenario - training models on data from the targeted user base - presents significant privacy concerns. To address both challenges, we propose to fuse image SR with federated learning, allowing real-world degradations to be directly learned from users without invading their …

abstract arxiv blind cs.ai cs.cv cs.et cs.lg current data eess.iv federated learning image leads privacy research resolution training training models type user data world