Emotion Recognition and understanding for Emotional Human-Robot Interaction Systems

Preface
Contents
1 Introduction
1.1 Emotion Feature Extraction and Recognition
1.2 Emotion Understanding
1.3 Emotional Human-Robot Interaction System
1.4 Organization of This Book
References
2 Multi-modal Emotion Feature Extraction
2.1 Introduction
2.2 Facial Expression Feature Extraction
2.3 Speech Emotion Feature Extraction
2.4 Gesture Feature Extraction
2.5 Summary
References
3 Deep Sparse Autoencoder Network for Facial Emotion Recognition
3.1 Introduction
3.2 Softmax Regression Based Deep Sparse Autoencoder Network
3.3 ROI Based Face Image Preprocessing
3.4 Expand the Encode and Decode Network
3.5 Softmax Regression
3.6 Overall Weight Training
3.7 Experiments
3.7.1 Fine-Tune Effect on Performance of Recognition
3.7.2 The Number of Hidden Layer Node's Effect on Performance of Recognition
3.7.3 Recognition Rate
3.8 Summary
References
4 AdaBoost-KNN with Direct Optimization for Dynamic Emotion Recognition
4.1 Introduction
4.2 Dynamic Feature Extraction Using Candide-3 Model
4.3 Adaptive Feature Selection Based on Plus-L Minus-R Selection
4.4 AdaBoost-KNN Based Emotion Recognition
4.5 AdaBoost-KNN with Direct Optimization for Emotion Recognition
4.6 Experiments
4.6.1 Experimental Environment and Data Selection
4.6.2 Simulations and Analysis
4.6.3 Preliminary Application Experiments on Emotional Social Robot System
4.7 Summary
References
5 Weight-Adapted Convolution Neural Network for Facial Expression Recognition
5.1 Introduction
5.2 Facial Expression Image Preprocessing
5.3 Principal Component Analysis for Extracting Expression Feature
5.4 Weight-Adapted Convolution Neural Network for Recognizing …
5.4.1 Feature Learning Based on Deep Convolution Neural Network
5.4.2 Softmax Regression for Feature Recognition
5.5 Hybrid Genetic Algorithm for Optimizing Weight Adaptively
5.6 Experiments
5.7 Summary
References
6 Two-Layer Fuzzy Multiple Random Forest for Speech Emotion Recognition
6.1 Introduction
6.2 Feature Extraction
6.3 Fuzzy-c-Means Based Features Classification
6.4 Two-Layer Fuzzy Multiple Random Forest
6.5 Experiments
6.5.1 Data Setting
6.5.2 Environment Setting
6.5.3 Simulations and Analysis
6.6 Summary
References
7 Two-Stage Fuzzy Fusion Based-Convolution Neural Network for Dynamic Emotion Recognition
7.1 Introduction
7.2 Dynamic Emotion Feature Extraction
7.3 Deep Convolution Neural Network for Extracting High-Level Emotion Semantic Features
7.4 Feature Fusion Based on Canonical Correlation Analysis
7.5 Decision Fusion Based on Fuzzy Broad Learning System
7.6 Two-Stage Fuzzy Fusion Strategy
7.7 Experiments
7.7.1 Data Setting
7.7.2 Experiments for Hyperparameters
7.7.3 Experimental Results and Analysis
7.8 Summary
References
8 Multi-support Vector Machine Based Dempster-Shafer Theory for Gesture Intention Understanding
8.1 Introduction
8.2 Foreground Segmentation and Feature Extraction
8.2.1 Foreground Segmentation Based on Depth and RGB Images
8.2.2 Speeded-Up Robust Features Based Gesture Feature Extraction
8.3 Encoding Speeded-Up Robust Features: Sparse Coding
8.4 Multi-class Linear Support Vector Machines
8.5 Dempster-Shafer Evidence Theory for Decision-Level Fusion
8.6 Experiments
8.6.1 Experimental Setting
8.6.2 Experimental Environment and Setup
8.6.3 Experimental Results and Analysis
8.7 Summary
References
9 Three-Layer Weighted Fuzzy Support Vector Regressions for Emotional Intention Understanding
9.1 Introduction
9.2 Support Vector Regression
9.3 Three-Layer Fuzzy Support Vector Regression
9.4 Characteristics Analysis of Emotional Intention Understanding
9.5 Three-Layer Fuzzy Support Vector Regression-Based Intention Understanding Model
9.6 Experiments
9.6.1 Experiment Setting
9.6.2 Experiments on Three-Layer Fuzzy Support Vector Regression Based Intention Understanding Model
9.7 Summary
References
10 Dynamic Emotion Understanding Based on Two-Layer Fuzzy Fuzzy Support Vector Regression-Takagi-Sugeno Model
10.1 Introduction
10.2 Dynamic Emotion Recognition Using Candide3-Based …
10.3 Two-Layer Fuzzy Support Vector Regression for Emotional …
10.4 Two-Layer Fuzzy Support Vector Regression Takagi-Sugeno Model for Emotional Intention Understanding
10.5 Experiments
10.5.1 Experimental Environment
10.5.2 Self-Built Data
10.5.3 Experiments on Dynamic Emotion Recognition and Understanding
10.6 Summary
References
11 Emotion-Age-Gender-Nationality Based Intention Understanding Using Two-Layer Fuzzy Support Vector Regression
11.1 Introduction
11.2 Two-Layer Fuzzy Support Vector Regression
11.2.1 Support Vector Regression
11.2.2 Two-Layer Fuzzy Support Vector Regression
11.3 Intention Understanding
11.3.1 Emotion Based Intention Understanding
11.3.2 Characteristics Analysis
11.4 Intention Understanding Model
11.4.1 Emotion Recognition
11.4.2 Questionnaire
11.4.3 ID Mapping
11.5 Two-Layer Fuzzy Support Vector Regression Based Intention Understanding
11.5.1 Intention Generation by Fuzzy Inference
11.6 Memory Retrieval for Intention Understanding
11.7 Experiments
11.7.1 Experiment Setting
11.7.2 Experiments on Two-Layer Fuzzy Support Vector Regression Based Intention Understanding Model
11.8 Summary
References
12 Emotional Human-Robot Interaction Systems
12.1 Introduction
12.2 Basic Emotional Human-Robot Interaction Systems
12.3 Design of Emotional Human-Robot Interaction System
12.4 Summary
References
13 Experiments and Applications of Emotional Human-Robot Interaction Systems
13.1 Introduction
13.2 Emotional Interaction Scenario Setting
13.3 Multi-modal Emotion Recognition Experiments Based on Facial …
13.4 Emotional Intention Understanding Experiments Based …
13.5 Application of Multi-modal Emotional Intention Understanding
13.5.1 Self-built Data
13.5.2 Experiments on Dynamic Emotion Recognition and Understanding
13.6 Summary
References
Index