An automatic 3D face model segmentation for acquiring weight motion area

Abstract

Inside facial animation works there is an animator that need to be skilled enough to produce detailed animation, so the facial animation can be smooth when doing facial expressions. Every animated character requires special handling based on the characteristics of the size and location of the bone. This process, where every face model need special handling were time consuming and tedious work. For that issue this research propose method for using motion capture marker data in 3D face model for automatically segment weight motion area based on the feature point. Marker data that came from motion capture of human model will be used to represent a centroid of vertex cluster that forming expressions in animated character. The data grouping process will be spherical coordinate result calculation between feature point and vertices using modified nearest neighbor algorithm. The result obtained in this research will show the weight motion area that generated automatically from the feature point based on nearest neighbor algorithm in a 3D face model.