1 paper · 1 filter
Jingwen Guo, Hong Liu, Shitong Sun +3
Existing skeleton-based action recognition methods typically follow a centralized learning paradigm, which can pose privacy concerns when exposing human-related videos. Federated L…