Gamasutra

Cloth

Tuesday, 26 July, 9:00 am - 10:30 am, Anaheim Convention Center, Ballroom C
Session Chair: Derek Nowrouzezahrai, Université de Montréal

A Compiler for 3D Machine Knitting

This compiler can translate high-level descriptions of knitted 3D objects into machine instructions for control of an industrial knitting machine.

James McCann
Carnegie Mellon University, The Walt Disney Company

Lea Albaugh
The Walt Disney Company, Disney Research

Vidya Narayanan
The Walt Disney Company, Disney Research

April Grow
University of California, Santa Cruz, The Walt Disney Company, Disney Research

Wojciech Matusik
Massachusetts Institute of Technology

Jennifer Mankoff
Disney Research Pittsburgh, Carnegie Mellon University

Jessica Hodgins
Carnegie Mellon University, The Walt Disney Company, Disney Research

Physics-Driven Pattern Adjustment for Direct 3D Garment Editing

To modify garment designs, which designers envision in 3D, existing tools require manual editing of 2D patterns. This paper proposes a framework that allows designers to directly apply the changes they seek in 3D and creates 2D patterns that replicate this envisioned geometry when lifted into 3D via physical simulation.

Aric Bartle
Stanford University

Vladimir Kim
Adobe Research

Danny Kaufman
Adobe Research

Alla Sheffer
The University Of British Columbia

Floraine Berthouzoz
Adobe Research

Nicholas Vining
The University Of British Columbia

Matching Real Fabrics With Micro-Appearance Models

This paper presents a framework for fitting appearance-model parameters to measurements and applies it to a study of the effectiveness of several light-scattering models and geometric representations used in fabric modeling.

Pramook Khungurn
Cornell University

Daniel Schroeder
Google

Shuang Zhao
University of California, Irvine

Kavita Bala
Cornell University

Steve Marshier
Cornell University

Fitting Procedural Yarn Models for Realistic Cloth Rendering

This approach fits CT measurements of physical yarns to procedural models automatically and recovers a full range of statistical parameters. Compared to existing micro-appearance fabric models, which usually involve large data sets, this fitted procedural representation is compact and easy to edit, yet offers similar levels of virtual realism.

Shuang Zhao
University of California, Irvine

Fujun Luan
Cornell University

Kavita Bala
Cornell University