Sketch-based instancing of parameterized 3D models

3Citations
Citations of this article
3Readers
Mendeley users who have this article in their library.
Get full text

Abstract

We present a system for constructing 3D models from simple hand-drawn sketches. The system exploits a priori knowledge about the type of object being sketched. In particular, we assume that the class of object being drawn is known, e.g., a rocket ship, and that the object will be drawn using a fixed number of known components. By using these assumptions, we show that we can build a sketch-based modeling system that, while object specific, is capable of quickly creating geometric models that are beyond the scope of current sketch-based modeling approaches. Key to our approach is the use of a K-means classifier for labeling each drawn stroke as being a particular component in the generic model. We demonstrate our approach by applying this classifier to face and rocket sketches. © 2010 Springer-Verlag Berlin Heidelberg.

Cite

CITATION STYLE

APA

Xiao, D., Pan, Z., & Zhou, R. (2010). Sketch-based instancing of parameterized 3D models. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6249 LNCS, pp. 550–561). https://doi.org/10.1007/978-3-642-14533-9_56

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free