Date: March 19th, 2009
Room: DMP 310
Speaker: Alla Sheffer
Title: Shape Understanding and Modeling

Abstract:

Shape is one of the most prominent characteristics of the objects surrounding us and one which is surprisingly non-trivial to describe algorithmically. In this talk I'll present some recent advances in modeling, analysis and capture of 3D shapes of everyday objects, specifically focusing on garment capture and shape orientation.

Garments are one example of a class of objects ubiquitous in our everyday surroundings but surprisingly challenging to model or even describe. Due to their ubiquity realistic static and dynamic garments are a critical component of character animation. In this talk I'll describe some novel approaches for modeling and capture of garments, including a method for modeling of believable dynamic folds and wrinkles.

Humans usually associate an upright orientation with objects, placing them in a way that they are most commonly seen in our surroundings. Using the famous "shape ever follows function" paradigm I'll demonstrate that it is often indeed possible to infer a model's upright orientation by analyzing its geometry.