brett allen 1,2 , brian curless 1 , zoran popovi ć 1 , and aaron hertzmann 3

24
Learning a correlated model of identity and pose- dependent body shape variation for real-time synthesis Brett Allen 1,2 , Brian Curless 1 , Zoran Popović 1 , and Aaron Hertzmann 3 1 University of Washington 2 Industrial Light & Magic 3 University of Toronto

Upload: sezja

Post on 12-Jan-2016

32 views

Category:

Documents


0 download

DESCRIPTION

Learning a correlated model of identity and pose-dependent body shape variation for real-time synthesis. Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3 1 University of Washington 2 Industrial Light & Magic 3 University of Toronto. movies. games. telepresence. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Learning a correlated model of identity and pose-dependent body shape variation

for real-time synthesis

Brett Allen1,2, Brian Curless1, Zoran Popović1, and Aaron Hertzmann3

1 University of Washington2 Industrial Light & Magic

3 University of Toronto

Page 2: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Motivation

movies games

telepresence design

Page 3: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Goal

• We would like to be able to generate body models of any individual in any pose.

Identity +

pose shape

- want to synthesize models in real-time

- model should be learnable from real data

Page 4: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Data

• CAESAR data set: 44 subjects in 2 poses• Multi-pose data set: 5 subjects in 16 poses• Dense-pose data set: 1 subject in 69 poses

[Anguelov et al. 2005]

Page 5: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Related work

Page 6: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Anatomical methods

Aubel 2002

Chadwick et al. 1989Turner and Thalmann 1993Scheepers et al. 1997Wilhelms and Van Gelder 1997…

Page 7: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Example-based methods

7000

7000

7000

0

0

0

z

y

x

z

y

x

v

),( Φpv f

v = shape vectorp = example parameters = function parameters

Given: n examples v0…vn and n sets of parameters p0…pn (optional)find .

Page 8: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Scattered data interpolation

Allen et al. 2002

Lewis et al. 2000Sloan et al. 2001Kry et al. 2002

i

iif φΦp ),(columns of = key shapesi = reconstruction weights from applying k-nearest neighbors or RBFs on p

1

2

43

1

3

2

4

Page 9: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

An aside on enveloping

Enveloping + scattered data interpolation= “corrective enveloping”

Page 10: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Latent variable modelsvWxvWx }),{,(f

x = latent variable (component weights)W = components in columnsv = average shape

Blanz & Vetter 1999

Allen et al 2003Seo et al 2003Anguelov et al 2005…

Page 11: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Pose variation vs body variation

Sloan et al. 2001

Page 12: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Pose variation vs body variation

Anguelov et al. 2005

KωvWxv

Page 13: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Our approach

iii

k

φv

b

φ

φ

c

cWxc

1

c = “character vector”: all information needed to put a character in any posev = shape in a particular pose

Intrinsic skeletonparameters: bonelengths and carrying angles

Page 14: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Two Problems

1. Scans might not be at “key” poses

2. Scans are not complete

Maximize: p(c | {e()})

…actually, we don’t know the pose or skinning weights either:

Maximize: p(c, s, q{} | {e()})

Page 15: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Maximum a posteriori estimation

})({log)(log)(log

2

1)2log(5.1log

})({)()(),,|(

}){|}{,,(

1 1

2)()(2

2

1 1

)(

)(

qsc

ve

qscqsc

eqsc

v

v

vvv

ppp

nnP

pppepP

pP

n n

iii

n n

ii

Page 16: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Results

Page 17: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Going to multiple characters

• One possibility: Learn several character vectors separately, then run PCA.

• Two problems:– the character vector contains values that have

different scales (rest positions, offsets, bone lengths)

– we don’t have enough data!

Page 18: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Identity variation

),0();,(),,( 2),(

NNg ii I0xqsce

cWxc

~ ~

c is the character vector of the th example persong(c,s,q) applies skinning and pose space deformations

Page 19: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Alternating optimization

• We initialize the {x} with the weights from running PCA on estimated skeleton parameters.

• We then optimize W, c, s, q.

• Then we optimize for {x}.

• Repeat…

Page 20: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Results

Page 21: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Results

(video)

Page 22: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Conclusions

We present a flexible approach for learning body shape variation between individuals and between poses, including the interrelationship between the two.+ very general: can handle irregular and incomplete sampling in regard to both the poses/identities scanned, and in the surfaces themselves+ the learned model can generate body shapes very quickly (over 75 fps)

Page 23: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Limitations

• You need a lot of data! Our data set was too sparse in some areas.

• Some poses are hard to capture.• It’s very hard to compensate for the

skinning artifacts.• The shape matching could be improved

(high-frequency details are lost if the matching is poor).

Page 24: Brett Allen 1,2 , Brian Curless 1 , Zoran Popovi ć 1 , and Aaron Hertzmann 3

Acknowledgements

• UW Animation Research Labs• Washington Research Foundation• National Science Foundation, NSERC, CFI• Microsoft Research, Electronic Arts, Sony, Pixar• Kathleen Robinette and the AFRL lab• Dragomir Anguelov• Domi Pitturo