Facial expression classificationIn still imagesAngel Gutiérrez, Montse Pardàs
IntroductionFace detectionContour detectionExpression estimationSystem implementation - DEAResultsConclusions
INTRODUCTION· Basic expressions (Ekman i Friesen – 1971)   Joy        Anger       Neutral      Sad     Surprise
INTRODUCTIONObjective      • To develop an automatic system which can      recognize a facial expression in a still image....
INTRODUCTIONGeneric scheme. Methods.   INPUT              FACE               INFORMATION         EXPRESSION           FACI...
FACE      FEATURES    EXPRESSIONDETECTION   DETECTION    DECISION
FACE DETECTIONViola and Jones:     · Fast search     · Robust to background     · Uses local texture features     · Trains...
FACIAL FEATURE CONTOUR DETECTION            Model based method          Active Shape Models        Active Appearance Models
FACIAL FEATURE CONTOUR DETECTIONBased on Active Appearance Model softwareimplemented by Stegmann*  * M. B. Stegmann, B. K....
FACIAL EXPRESSION             INPUT          CLASSIFIER         KNOWN             DATA            SYSTEM            OUTPUT...
RESULTSDatabase  · 192 labeled images.
RESULTS
RESULTSTEST 1       FACE               FEATURES          EXPRESSION     DETECTION            DETECTION          DECISION  ...
RESULTSTEST 2        FACE                FEATURES          EXPRESSION      DETECTION             DETECTION          DECISI...
CONCLUSIONS· Automatic system for facial expression detection in 3 stages:                                FACIAL FEATURE  ...
Upcoming SlideShare
Loading in …5
×

Facial expressionclass barca

372 views
319 views

Published on

Published in: Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
372
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
16
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Facial expressionclass barca

  1. 1. Facial expression classificationIn still imagesAngel Gutiérrez, Montse Pardàs
  2. 2. IntroductionFace detectionContour detectionExpression estimationSystem implementation - DEAResultsConclusions
  3. 3. INTRODUCTION· Basic expressions (Ekman i Friesen – 1971) Joy Anger Neutral Sad Surprise
  4. 4. INTRODUCTIONObjective • To develop an automatic system which can recognize a facial expression in a still image.Applications • Monitor for drivers • Stress detection • Facial coder • Entertainment/ Games • etc.
  5. 5. INTRODUCTIONGeneric scheme. Methods. INPUT FACE INFORMATION EXPRESSION FACIAL DETECTION EXTRACTION DECISION EXPRESSION IMAGE· Viola and Jones · Gabor Wavelets filters · Hidden Markov Models · Active appearance models · Neural Networks· Region based · Dense motion fields · Probabilistic models · Feature points tracking
  6. 6. FACE FEATURES EXPRESSIONDETECTION DETECTION DECISION
  7. 7. FACE DETECTIONViola and Jones: · Fast search · Robust to background · Uses local texture features · Trains classifiers for face/non-face classes · Uses a cascade of classifiers structure (Adaboost)Region-based: · Refines the face detection to obtain a better initizialization
  8. 8. FACIAL FEATURE CONTOUR DETECTION Model based method Active Shape Models Active Appearance Models
  9. 9. FACIAL FEATURE CONTOUR DETECTIONBased on Active Appearance Model softwareimplemented by Stegmann* * M. B. Stegmann, B. K. Ersbøll, R. Larsen, FAME - A Flexible Appearance Modelling Environment, IEEE Transactions on Medical Imaging, vol. 22(10), pp. 1319-1331, Institute of Electrical and Electronics Engineers (IEEE), 2003 - Facial feature contours are represented by a 58 points model
  10. 10. FACIAL EXPRESSION INPUT CLASSIFIER KNOWN DATA SYSTEM OUTPUT Class 1, Class 2, LABELED ... DATABASE Class N. Bayesian framework with probabilistic model: Mixture of multivariate gaussians, trained with EM for each class
  11. 11. RESULTSDatabase · 192 labeled images.
  12. 12. RESULTS
  13. 13. RESULTSTEST 1 FACE FEATURES EXPRESSION DETECTION DETECTION DECISION 95,47% H A N Su Sa H 96% 0% 0% 4% 0% A 0% 98% 0% 0% 2% N 0% 0% 94% 6% 0% 95,47 % Su 4% 0% 3% 93% 0% Sa 0% 0% 3% 0% 97%
  14. 14. RESULTSTEST 2 FACE FEATURES EXPRESSION DETECTION DETECTION DECISION 84,66% H A N Su Sa H 96% 0% 2% 2% 0% A 6% 79% 4% 4% 8% N 0% 3% 69% 13% 16% 84,66 % Su 0% 0% 4% 96% 0% Sa 0% 3% 14% 0% 83%
  15. 15. CONCLUSIONS· Automatic system for facial expression detection in 3 stages: FACIAL FEATURE EXPRESSION FACE DETECTION CLASSIFICATION CONTOURS DET.·Correct classification rate 85 %, with 5 classes.· Only frontal faces. Problems with facial hair and sometimes with glasses

×