Title: multimodal%20emotion%20recognition
1multimodal emotion recognition
- recognition models- application dependency
- discrete / dimensional / appraisal theory models
- theoretical models of multimodal integration
- direct / separate / dominant / motor integration
- modality synchronization
- visemes/ EMGs FAPs / SC-RSP speech
- temporal evolution and modality sequentiality
- multimodal recognition techniques
- classifiers context goals
cognition/attention modality significance in
interaction
2multimodal emotion recognition
- Two approaches have been developed and used for
audiovisual emotion recognition - Separated Recognition
- An attention-feedback recurrent neural
network applied to emotion recognition from
speech. - A neurofuzzy system including a-priori
knowledge used for emotion recognition from
facial expressions.
3Separated Recognition
- The goal was to evaluate performance of the
obtained recognition by each modality. Visual
feeltracing was required. - Pause detection tune-based analysis, with
speech playing the main role, was the means to
synchronise the two modalities.
4Emotion analysis facial expressions
- A rule-based system for emotion recognition was
created, characterising a users emotional state
in terms of the six universal, or archetypal,
expressions (joy, surprise, fear, anger, disgust,
sadness. - Rules have been created in terms of the MPEG-4
FAPs for each of these expressions.
5Sample Profiles of Anger
A1 F422, 124, F31-131, -25, F32-136,-34,
F33-189,-109, F34-183,-105, F35-101,-31,
F36-108,-32, F3729,85, F3827,89 A2
F19-330,-200, F20-335,-205, F21200,330,
F22205,335, F31-200,-80, F32-194,-74,
F33-190,-70, F34-190,-70 A3 F19
-330,-200, F20-335,-205, F21200,330,
F22205,335, F31-200,-80, F32-194,-74,
F3370,190, F3470,190
6Emotion analysis facial expressions
G the value of a corresponding FAP
f Values derived from the calculated distances
7(No Transcript)
8Expressions Rule more often activated (
examined cases)
- Anger open_jaw_low, lower_top_midlip_medium,
raise_bottom_midlip_high, raise_left_inner_eyebrow
_low, raise_right_inner_eyebrow_low,
raise_left_medium_eyebrow_low, raise_right_medium_
eyebrow_low, squeeze_left_eyebrow_high,
squeeze_right_eyebrow_high, wrinkles_between_eyebr
ows_high, raise_left_outer_cornerlip_medium,
raise_right_outer_cornerlip_medium (47) - Joy open_jaw_high, lower_top_midlip_low,
raise_bottom_midlip_verylow, widening_mouth_high,
close_left_eye_high, close_right_eye_high (39) - Disgust open_jaw_low, lower_top_midlip_low,
raise_bottom_midlip_high, widening_mouth_low,
close_left_eye_high, close_right_eye_high,
raise_left_inner_eyebrow_medium,
raise_right_inner_eyebrow_medium,
raise_left_medium_eyebrow_medium,
raise_right_medium_eyebrow_medium,
wrinkles_between_eyebrows_medium 33)
9Expressions Rule more often activated (
examined cases)
- Surprise open_jaw_high, raise_bottom_midlip_veryl
ow, widening_mouth_low, close_left_eye_low,
close_right_eye_low, raise_left_inner_eyebrow_high
, raise_right_inner_eyebrow_high,
raise_left_medium_eyebrow_high,
raise_right_medium_eyebrow_high,
raise_left_outer_eyebrow_high, raise_right_outer_e
yebrow_high, squeeze_left_eyebrow_low,
squeeze_right_eyebrow_low, wrinkles_between_eyebro
ws_low (71) - Neutral open_jaw_low, lower_top_midlip_medium,
raise_left_inner_eyebrow_medium,
raise_right_inner_eyebrow_medium,
raise_left_medium_eyebrow_medium,
raise_right_medium_eyebrow_medium,
raise_left_outer_eyebrow_medium,
raise_right_outer_eyebrow_medium,
squeeze_left_eyebrow_medium, squeeze_right_eyebrow
_medium, wrinkles_between_eyebrows_medium,
raise_left_outer_cornerlip_medium,
raise_right_outer_cornerlip_medium (70)
10Expression based Emotion Analysis Results
- These rules were extended to deal with 2-D
continuous (activation-evaluation) 4 quadrant
emotional space - They were applied to QUB SALAS generated data to
test the performance to real life emotion
expressive data sets.
11(No Transcript)
12(No Transcript)
13Clustering/Neurofuzzy Analysis of Facial Features
- The rule-based expression/emotion analysis system
was extended to handle specific characteristics
of each user in continuous 2-D emotional
analysis. - Novel clustering and fuzzy reasoning techniques
were developed and used for producing specific
FAP ranges (around 10 clusters) for each user and
providing rules to handle them. - Results on the continuous 2-D emotional framework
with SALAS data indicate that a good performance
(reaching 80) was obtained applying the adapted
systems to each specific user.
14Direct Multimodal Recognition
- The attention-feedback recurrent neural network
architecture (ANNA) was applied to emotion
recognition based on all input modalities. - Features extracted from all input modalities
(linguistic, paralinguistic speech, FAPs) were
provided by processing and analysing common SALAS
emotional expressive data.
15Emotion Recognition based on ANNA
- ANNA hidden layer emotion state, feedback
control for attention ( IMC) - Learning laws for ANNA developed
- ANNA fuses all modalities or only one
16BASIC EMOTION RECOGNITION ARCHITECTURE
Feature vector Inputs
Attention control system
Output as recognised emotional state
?
?
Emotion state as hidden layer
17Text Post-Processing Module
- Prof. Whissell compiled
- Dictionary of Affect in Language (DAL)
- Mapping of 9000 words ? (activation-evaluation),
based on students assessment - Take words from meaningful segments obtained by
pause detection ? (activation-evaluation) space - But humans use context to assign emotional
content to words
18ANNA on top correlated ASSESS features
- Quadrant match using top 10 activation features
top 10 evaluation features and activation
evaluation output space
Feeltracer jd cc dr em
Avg Quad Match 0.42 0.39 0.37 0.45
Std Dev 0.03 0.02 0.02 0.04
19ANNA on top correlated ASSESS features
- Half-plane match using top 10 activation features
and activation only output space
Feeltracer jd cc dr em
Avg Quad Match 0.75 0.66 0.64 0.74
Std Dev 0.02 0.02 0.02 0.03
20Multi-modal Results
- 500 training epochs, 3 runs per dataset, final
results being averaged (with associated Sdev). - 5 hidden layer (EMOT) neurons and 5 feedback
layer (IMC) neurons learning rate fixed at
0.001. - Of each dataset 4 parts used for training and 1
part for testing the net on unseen inputs. - AActivation, EEvaluation FT stands for the
FeelTracer used as supervisor AVG denotes the
average quadrant match (for 2D-space) or average
half-plane match for (1D-space) over 3 runs. - PCA on the ASSESS features to reduce them from
about 500 to around 7-10 as describing most of
the volatility.
21Multi-modal Results Using A Output only
- Classification using A output only relatively
high - (in three cases up to 98, and with two more at
90 or above) - Effectiveness of data from FeelTracer EM
- Average success rates of (86, 88, 98, 89,
95, 98, 98) for 7 choices of input
combinations (ass/fap/dal/af/df/ad/adf) - Also high success with Feeltracer JD
- Consistently lower values for FeelTracer DR
- (all in 60-66 band)
- Also for CC (64, 54, 75, 63, 73, 73, 73).
22Ontology Representation Facial
Expression/Emotion Analysis
- Use ontologies for real life usage of facial
expression/emotion analysis results - Extensibility
- (Ontologies form an excellent basis for
considering issues like constrained reasoning,
personalisation, adaptation, which have been
shown crucial for applying our results to real
life applications ) - Standardisation
- (OWL ontologies form a standard knowledge
representation and reasoning Web framework)
23Facial Emotion Analysis Ontology Development
- An ontology has been created to represent the
geometry and different variations of facial
expressions based on the MPEG-4 Face Animation
Parameters (FDPs) and Face Definition Parameters
(FAPs). - The ontology was built using the ontology
language OWL DL and the Protégé OWL ontologies
development Plugin. - The ontology will be the tool for extending the
obtained results to real life applications
dealing with specific users profiles
constraints.
24Concept and Relation Examples
-
- Concepts
- Face
- Face_Animation_Parameter
- Face_Definition_Parameter
- Facial_Expression
- Relations
- is_Defined_By
- is_Animated_By
- has_Facial_Expression
-