Probabilistic Model for Virtual Garment Modeling - Advances in Image and Graphics Technologies - page 203

Probabilistic Model for Virtual Garment Modeling

Shan Zeng, Fan Zhou, Ruomei Wang, and Xiaonan Luo

National Engineering Research Center of Digital Life, State-Province Joint Laboratory of
Digital Home Interactive Applications, School of Information Science Technology,
Sun Yat-sen University, Guangzhou 510006, China
[e-mail]

Abstract. Designing 3D garments is difficult, especially when the user lacks
professional knowledge of garment design. Inspired by the assemble modeling,
we facilitate 3D garment modeling by combining parts extracted from a
database containing a large collection of garment component. A key challenge
in assembly-based garment modeling is the identifying the relevant components
that needs to be presented to the user. In this paper, we propose a virtual
garment modeling method based on probabilistic model. We learn a
probabilistic graphic model that encodes the semantic relationship among
garment components from garment images. During the garment design process,
the Bayesian graphic model is used to demonstrate the garment components that
are semantically compatible with the existing model. And we also propose a
new part stitching method for garment components. Our experiments indicates
that the learned Bayesian graphic model increase the relevance of presented
components and the part stitching method generates good results.

Keywords: Garment modeling, Bayesian Probabilistic Graphic Model, and Part
Stitching.

1 Introduction

Modeling garments is essential for virtual fitting and can benefit other applications
such as films and games. The modeling garment is a challenge task because the
garment is complex in structure: it contains different kinds of components and varies
of garment style. Berthouzoz [1] created detailed garment by seaming 2D patterns
automatically. Bradley, Li, Zhou [2,3,4] model the garment from photos, it relaxes the
professional requirement of garment design. However, their model couldn't model
detailed garment that contains pocket, belt, and button, etc.
A successful 3D garment design system should be simple, intuitive to use, and
provides multiple design options for different design intent. Assembly-based
modeling provides a promising new approachto 3D garment modeling. If we have a
database of garment components, we can model garments by assembling the
components. Identification of relevant components to be presented to the user is a key
problem in assembly-based modeling [5,6,7]. The advantage of assembly-based
modeling is that users do not need to design 2D patterns.