Exploring the space of human body shapes: data-driven synthesis under anthropometric control

1
Exploring the space of human body
shapesdata-driven synthesisunder
anthropometric control

• Brett Allen
• Brian Curless
• Zoran Popovic
• University of Washington

2004-01-2188
2
Motivation

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

3
Motivation

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

4
CAESAR

• Civilian American European Surface
  Anthropometry Resource

• thousands of subjects in the U.S. and Europe
• traditional anthropometry
• demographic survey
• laser range scans

Well use 250 of these scans (125 male, 125
female).
5
Scan detail

• 250,000 triangles
• incomplete coverage

6
Overview

• 1. Introduction
• 2. Building a model
• 3. Synthesis editing

7
Overview

• 1. Introduction
• 2. Building a model
• 3. Synthesis editing

8
The Correspondence Problem
9
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
scan
template
10
Matching algorithm
11
Overview

• 1. Introduction
• 2. Building a model
• 3. Synthesis editing

12
Statistical analysis
mean PCA component 1
13
Statistical analysis
mean PCA component 2
14
Statistical analysis
mean PCA component 3
15
PCA reconstruction
16
Fitting to attributes

• We can correlate the PCA reconstructions of
  our scanned people with known attributes

17
Fitting to attributes
18
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
19
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

20
Future work

• - analyze shape variation between poses

21
Future work

• - combine with anatomical models and physical
  simulation

Aubel 2003
22
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