Large-sized objects such as human manikins need a proper a priori mesh segmentation process for successful 3D printing. This paper applied well-known statistical mesh clustering functions, including k-means, k-medoids, DBSCAN, and pairwise distance. Especially, an inventive clustering metric, designated as point-to-bone distance, was proposed to take advantage of bone structure which can be acquired easily from free software such as Adobe Mixamo. Furthermore, the cut parts were oriented to the optimal directions, in which the minimal support structure was expected, using our previous work. Now any textile or apparel scientists can easily 3D-print their human manikins with arbitrary shapes and sizes with the help of python language.