Automatic Single-View Character Model Reconstruction

Philip Buchanan
University of Canterbury

R. Mukundan
University of Canterbury

Michael Doggett
Lund University

Expressive 2013 (SBIM), July 19-20 2013, Anaheim, U.S.A.

Abstract

In this paper we present a new method for automatically construct- ing 3D meshes from a single input image. With the increasing con- tent demands of modern digital entertainment and the expectation of involvement from users, automatic artist-free systems are an im- portant step in allowing user generated content and rapid game pro- totyping. Our system proposes a novel heuristic for the creation of a 3D mesh from a single piece of non-occluding 2D concept art. By extracting a skeleton structure, approximating the 3D ori- entation and analysing line curvature properties, appropriate cen- trepoints can be found around which to create the cross-sectional slices used to build a final triangle mesh. Our results show that a single 2D input image can be used to generate a rigged 3D low- polygon model suitable for use in realtime applications.

Downloads

The paper. [pdf 5.8 MB]