User-Defined Machining Feature Recognition Based on Semantic Reasoning for B-rep Models

Abstract

Extracting wireframes from 3D meshes is a challenging problem. Existing methods are typically based on analyzing local shape properties, such as surface curvatures and angles between faces, which are generally sensitive to small features in the model and the complexity of the object. In this work, we propose a wireframe extraction method based on a geometric approximation of 3D meshes. Our idea is to utilize a well-established variational geometric partitioning method to derive a complete set of descriptive feature curves. The advantage of this strategy is that a reliable feature ltering mechanism can be inherently incorporated into the geometric approximation step, giving rise to descriptive wireframes that capture global structures of the 3D meshes. We experimented on various meshes. The results demonstrated that our method is superior to existing approaches in terms of the correctness and completeness of the extracted feature curves.