+ All Categories
Home > Documents > From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and...

From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and...

Date post: 31-Mar-2015
Category:
Upload: branden-pavey
View: 216 times
Download: 1 times
Share this document with a friend
Popular Tags:
18
From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and A.Raouzaiou, K.Karpouzis and S.Kollias S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University of Athens
Transcript
Page 1: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

From Facial Featuresto Facial Expressions

A.Raouzaiou, K.Karpouzis and S.KolliasA.Raouzaiou, K.Karpouzis and S.Kollias

Image, Video and Multimedia Systems LaboratoryNational Technical University of Athens

Page 2: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Outline

The concept of archetypal expressions

FAPs-based description and estimation of FAPs

Expression synthesis using profiles

Synthesis of intermediate emotions

Page 3: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Archetypal Expressions

Source: F. Parke and K. Waters, Computer Facial Animation, A K Peters

Also termed universaluniversal because they are recognized across cultures

Page 4: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Archetypal Expressions (cont.)Description of the

archetypal expressions through

muscle actionsTranslation of facial muscle movements

into FAPs

Creation of FAPs vocabulary for every

archetypal expression

Action Units (AUs) - FACS

raise_l_i_eyebrow

e.g.AU1= + raise_r_i_eyebrow

e.g. sadnessclose_t_l_eyelid, close_t_r_eyelid, close_b_l_eyelid, close_b_r_eyelid, raise_l_i_eyebrow, raise_r_i_eyebrow, raise_l_m_eyebrow, raise_r_m_eyebrow, raise_l_o_eyebrow, raise_r_o_eyebrow

Page 5: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

FAPs-based description

Discrete features offer a neat, symbolic representation of expressions

Not constrained to a specific face model Suitable for face cloning applications

MPEG-4 compatible Based on feature points, not complete features

Page 6: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

FAPs-based description (cont.)

Two issues should be addressed :

choice of FAPs involved in profiles’

formation definition of FAP intensities

Page 7: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Expression synthesis

Choice of FAPs is based on psychological data

Intensities are derived from expression database images

Page 8: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Estimation of FAPs

Absence of clear quantitative definition of FAPs

It is possible to model FAPs through FDP feature points movement using distances s(x,y)

e.g. close_t_r_eyelid (F20) - close_b_r_eyelid (F22) D13=s (3.2,3.4) f13= D13 - D13-NEUTRAL

Page 9: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Sample FAP vocabulary

Sadness:close_t_l_eyelid(F19), close_t_r_eyelid(F20 ), close_b_l_eyelid (F21), close_b_r_eyelid(F22), raise_l_i_eyebrow(F31), raise_r_i_eyebrow(F32 ), raise_l_m_eyebrow(F33), raise_r_m_eyebrow(F34), raise_l_o_eyebrow(F35), raise_r_o_eyebrow(F36)

Page 10: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Archetypal Expression Profiles

ProfileProfile: set of FAPs accompanied by the corresponding range of variation

definition of subsets of candidate FAPs

use of range variations obtained from statistics

animation of the corresponding profiles to

verify appropriateness

face formations from psychological studies

Page 11: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Sample Profiles of Anger

A1: F4[22, 124], F31[-131, -25], F32[-136,-34], F33[-189,-109],

F34[-183,-105], F35[-101,-31], F36[-108,-32], F37[29,85],

F38[27,89]

A2: F19[-330,-200], F20[-335,-205], F21[200,330],

F22[205,335], F31[-200,-80], F32[-194,-74], F33[-190,-70],

F34=[-190,-70]

A3: F19 [-330,-200], F20[-335,-205], F21[200,330],

F22[205,335], F31[-200,-80], F32[-194,-74], F33[70,190],

F34[70,190]

Page 12: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Emotion representation

Emotions can be approached as points on a plane defined by activation and evaluation

Page 13: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Intermediate Expression Profiles Same universal emotion category

Animation of the same FAPs using different intensitiesAbsence of expert knowledge for the (+, –) quadrant

worry < fear < terror

Page 14: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Intermediate Expression Profiles Different universal emotion categories

In the same evaluation half-planeAveraging of FAPs used in universal emotions

Page 15: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Intermediate Expression Profiles Different universal emotion categories

afraid + sad = depressed

Page 16: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Conclusions

FAPs provide a compact and established means of emotion representation

Necessary input from psychological and physiological studies

Universal emotions can be used to synthesize intermediate onesUseful for low-bitrate MPEG-4 applications

Page 17: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Extensions

Verification – Evaluation Initial resultsAcceptable performance for expression

grading Intermediate expressions: better results for

the negative evaluation half planeLack of linguistic rules for the (+, -) quadrant

Page 18: From Facial Features to Facial Expressions A.Raouzaiou, K.Karpouzis and S.Kollias Image, Video and Multimedia Systems Laboratory National Technical University.

Extensions

Personalized ECAsDetected facial feature points can be used to

adapt a generic ECA head (FDP FPs) Intermediate emotions based on processing

real data (FAP extraction)Processing real data temporal aspect of

FAPs


Recommended