Abstract
A direct and intuitive 3-D geometric operation interface using various-shaped deformable real objects as input tools is proposed. Elastic-conductive material block itself is used not only as a structural body but also as 3-D deformation sensor. To use various-shaped blocks as input tools other than simple parallelepiped ones, neural network is introduced to learn relationships between 3-D deformation of the blocks and measurements of their resistance change patterns. Also, a neural netwerk optimization algorithm is proposed for semi-automatic determination of the number and arrangement of wire terminals to measure 3-D deformation of the input tool. The measured and calculated deformation of the input tool is mapped to the object shape on a computer display using free form deformation. This interface enables users to use input tools of suitable shapes for their purposes, from a simple shape for general purposes like cubes to a specific shape like a doll figure.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call