Abstract
An efficient and stable skeletonization consisting of a Voronoi skeletonization following by a two step pruning is presented. The first pruning step removes Voronoi edges crossing the shape boundary. The second follows a Discrete Curve Evolution approach. Both pruning steps can be done very efficiently because entire Voronoi segments are pruned based on tests on points. The algorithm works in realtime and could be used in a gesture recognition interface for tabletop interfaces. © 2010 Springer-Verlag Berlin Heidelberg.
Cite
CITATION STYLE
Beristain, A., & Graña, M. (2010). A stable skeletonization for tabletop gesture recognition. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6016 LNCS, pp. 610–621). Springer Verlag. https://doi.org/10.1007/978-3-642-12156-2_46
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
