Exploring the space of human body shapes datadriven

Exploring the space of human body shapes: data-driven synthesis under anthropometric control Brett Allen Brian Curless Zoran Popović University of Washington 2004 -01 -2188

Motivation Traditional anthropometry has focused on sets of one-dimensional measurements.

Motivation Full body shape capture promises to advance the state of the art. ?

CAESAR Civilian American & European Surface Anthropometry Resource • • thousands of subjects in the U. S. and Europe traditional anthropometry demographic survey laser range scans We’ll use 250 of these scans (125 male, 125 female).

Scan detail ~250, 000 triangles incomplete coverage surface color 74 markers

Overview 1. Introduction 2. Building a model 3. Synthesis & editing

Overview 1. Introduction 2. Building a model 3. Synthesis & editing

The Correspondence Problem

Matching algorithm Find the shape that: 1. Matches the template markers to the scanned markers 2. Moves template vertices to scanned surface 3. Minimizes the deformation template scan

Matching algorithm

Overview 1. Introduction 2. Building a model 3. Synthesis & editing

Statistical analysis x 0 y 0 z 0 x 1 y 1 z 1 x 2 x 2 average male mean + PCA component #1

Statistical analysis x 0 y 0 z 0 x 1 y 1 z 1 x 2 x 2 average male mean + PCA component #2

Statistical analysis x 0 y 0 z 0 x 1 y 1 z 1 x 2 x 2 average male mean + PCA component #3

PCA reconstruction

Fitting to attributes We can correlate the PCA reconstructions of our scanned people with known attributes:

Fitting to attributes

Fitting to points Using the distribution of the PCA weights as a prior, we can find the most likely person that fits a set of point constraints. PCA variance user constraint optimized reconstruction

Summary Contributions: - an algorithm for creating a consistent mesh representation from range scan data. - several ways to explore the variation in human body shape, and to synthesize and edit body models

Future work - analyze shape variation between poses

Future work - combine with anatomical models and physical simulation + Aubel 2003

Acknowledgments - Kathleen Robinette and the CAESAR project Ethel Evans Domi Pitturo Daniel Wood - NSF NSERC Microsoft Research, Electronic Arts, Sony University of Washington Animation Research Labs 2004 -01 -2188