In this talk, we consider the Forward--Backward (FB) splitting algorithm and its variants (inertial schemes, FISTA, Tseng's FBF) for solving structured optimization problem. The goal of this talk is to establish the local convergence behavior of these methods when the involved functions are partly smooth relative to their active manifolds. We show that all these splitting methods correctly identify the active manifolds in a finite time, and then enter a local linear convergence regime, which we characterize precisely. This is illustrated by several numerical experiments.