In many geometry processing applications, the estimation of differential geometric quantities such as curvature or normal vector field is an essential step. When designing such estimators, we have to pay attention to both its theoretical properties and practical effectiveness. In this paper, we investigate a new class of estimators on digital shape boundaries based on Integral Invariants. More precisely, we provide proofs of multigrid convergence of curvature estimators which are easy to implement on digital data. Furthermore, we discuss about some algorithmic optimisations and detail a complete experimental evaluation.