000

001

002

003

004

005

006

007

008

009

010

011

012

013

014

015

016

017

018

019

020

021

022

023

024

025

026

027

028

029

030

031

032

033

034

035

036

037

038

039

040

041

042

043

044

000

001

002

003

004

005

006

007

008

009

010

011

012

013

014

015

016

017

018

019

020

021

022

023

024

025

026

027

028

029

030

031

032

033

034

035

036

037

038

039

040

041

042

043

044

ECCV

#366ECCV

#366

Total Moving Face ReconstructionSupplement Materials

Anonymous ECCV submission

Paper ID 366

We compare to single view method in Figure 1. Our method (outlined in red)obtains higher quality coarse as well as detail reconstruction due to estimationof 3D flow coupled with extreme non frontal pose and shading. State of theart single view methods do not account for 3D flow. Additionally, while weestimate an initial template (average shape) of the individual we would like toreconstruct, state of the art single view methods use models of other individualsas the template; this also decreases the quality of the results.

Input& Ours&Single/view&[30]&Average&Shapes&

Fig. 1. A comparison to single view method by Kemelmacher et al. [30].