Abstract
In this paper, we propose a practical algorithm for extracting curve skeletons from a 3D shape represented by a triangular mesh. We first construct an initial skeleton graph by copying the connectivity and geometry information from the input mesh. We then perform iterative skeletonization over the nodes of the skeleton graph using coupled processes of graph contraction and surface clustering. In the contraction step, the skeleton graph is simplified and regularized with surface clustering: mesh vertices are clustered, while the positions of nodes in the skeleton graph are updated at the same time. Eventually, the skeleton graph is automatically simplified to an approximately-centered curve skeleton. Our algorithm naturally produces a skeleton-to-surface mapping, making the output skeletons directly applicable to skinning deformation.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have