UAU-Net: Uncertainty-aware Representation Learning and Evidential Classification for Facial Action Unit Detection

ArXi:2604.21227v1 Announce Type: new Facial action unit (AU) detection remains challenging because it involves heterogeneous, AU-specific uncertainties arising at both the representation and decision stages. Recent methods have improved discriminative feature learning, but they often treat the AU representations as deterministic, overlooking uncertainty caused by visual noise, subject-dependent appearance variations, and ambiguous inter-AU relationships, all of which can substantially degrade robustness.