As an important information bridge connecting design and manufacturing, feature recognition has been intensively researched. However, most researches focused on regular shaped features. Unlike regular shaped features, freeform features are usually partitioned into faces and regions for the selection of proper machining tools and parameters to meet design specifications and improve machining efficiency. In this paper, a novel recognition method for freeform machining features combining both curvature-based segmentation and manufacturability analysis-based segmentation is developed. A hierarchical representation of freeform machining features is presented. A case study is presented to demonstrate the capability of the proposed method.