This book introduces readers to the latest exciting advances in human motion sensing and recognition, from the theoretical development of fuzzy approaches to their applications. The topics covered include human motion recognition in 2D and 3D, hand motion analysis with contact sensors, and vision-based view-invariant motion recognition, especially from the perspective of Fuzzy Qualitative techniques.
With the rapid development of technologies in microelectronics, computers, networks, and robotics over the last decade, increasing attention has been focused on human motion sensing and recognition in many emerging and active disciplines where human motions need to be automatically tracked, analyzed or understood, such as smart surveillance, intelligent human-computer interaction, robot motion learning, and interactive gaming. Current challenges mainly stem from the dynamic environment, data multi-modality, uncertain sensory information, and real-time issues.
These techniques are shown to effectively address the above challenges by bridging the gap between symbolic cognitive functions and numerical sensing & control tasks in intelligent systems. The book not only serves as a valuable reference source for researchers and professionals in the fields of computer vision and robotics, but will also benefit practitioners and graduates/postgraduates seeking advanced information on fuzzy techniques and their applications in motion analysis.

lgli/K:\!genesis\!repository9\spr\10.1007%2F978-3-662-53692-6.pdf

lgrsnf/K:\!genesis\!repository9\spr\10.1007%2F978-3-662-53692-6.pdf

nexusstc/Human Motion Sensing and Recognition: A Fuzzy Qualitative Approach/579ca7246c0a25061b57223647cab249.pdf

scihub/10.1007/978-3-662-53692-6.pdf

zlib/Computers/Honghai Liu, Zhaojie Ju, Xiaofei Ji, Chee Seng Chan, Mehdi Khoury (auth.)/Human Motion Sensing and Recognition: A Fuzzy Qualitative Approach_2943046.pdf

Hand Motion Recognition and Transfer: A Unified Framework for Human Hand Manipulation Recognition and Its Application

309157_Print.indd

Honghai Liu, (Professor of Intelligent Systems); Zhaojie Ju; Xiaofei Ji; Chee Seng Chan; Mehdi Khoury

Liu, Honghai, Ju, Zhaojie, Ji, Xiaofei, Chan, Chee Seng, Khoury, Mehdi

0009172

Springer Berlin Heidelberg : Imprint: Springer

Springer Spektrum. in Springer-Verlag GmbH

Steinkopff. in Springer-Verlag GmbH

Springer Nature

Studies in computational intelligence, volume 675, Berlin, Germany, 2017

Studies in Computational Intelligence, 675, Berlin, Heidelberg, 2017

Springer Nature, Berlin, Heidelberg, 2017

1st ed. 2017, 2017

Germany, Germany

2, 20170511

sm64676808

producers:
Acrobat Distiller 10.0.0 (Windows)

{"container_title":"Studies in Computational Intelligence","content":{"parsed_at":1698285731,"source_extension":"epub"},"edition":"1","isbns":["3662536900","3662536927","9783662536902","9783662536926"],"issns":["1860-949X","1860-9503"],"publisher":"Springer","series":"Studies in Computational Intelligence 675"}

Preface 6
Contents 9
Acronyms 14
1 Introduction 16
1.1 Human Motion Sensing Techniques 16
1.1.1 Vision-Based Sensing 17
1.1.2 Wearable Sensing 22
1.1.3 Multimodal Sensing 29
1.2 Recognition Approaches 30
1.2.1 Probabilistic Graphical Models 30
1.2.2 Support Vector Machines 32
1.2.3 Artificial Neural Networks and Deep Learning 33
1.2.4 Fuzzy Approaches 35
1.3 Challenges and Motivation 41
References 42
2 Fuzzy Qualitative Trigonometry 50
2.1 Introduction 50
2.2 Fuzzy Qualitative Trigonometry 52
2.3 Fuzzy Qualitative Trigonometric Functions 56
2.4 Derivatives of Fuzzy Qualitative Trigonometry 58
2.5 Conclusion 62
References 63
3 Fuzzy Qualitative Robot Kinematics 66
3.1 Introduction 66
3.2 Fuzzy Qualitative Robot Kinematics 67
3.2.1 Fuzzy Qualitative Transformations 67
3.2.2 Fuzzy Qualitative Robot Kinematics 70
3.3 Quantity Vector Aggregation 71
3.4 Case Study 74
3.5 Conclusion 79
References 80
4 Fuzzy Qualitative Human Motion Analysis 81
4.1 Introduction 81
4.2 FQ Human Motion Analysis 84
4.2.1 Human Body Modelling 84
4.2.2 Human Motion Tracking 85
4.2.3 Data Quantisation 87
4.2.4 Human Motion Representation 91
4.3 Experiments 95
4.3.1 Datasets and Preprocessing 95
4.3.2 Results and Analysis 98
4.3.3 Quantitative Comparison 102
4.3.4 Complexity Analysis 103
4.4 Conclusion 105
References 105
5 Fuzzy Gaussian Mixture Models 108
5.1 Introduction 108
5.2 Generalised GMMs 109
5.2.1 Conventional GMMs 110
5.2.2 Generalized Gaussian Models 111
5.2.3 EM Algorithm for Generalized GMMs 114
5.3 Fuzzy Gaussian Mixture Models 115
5.3.1 Fuzzy C-Means Clustering 116
5.3.2 Probability Based FGMMs 117
5.3.3 Distance Based FGMMs 118
5.3.4 Comparison of Probability/Distance Based FGMMs 119
5.4 Experiments 120
5.4.1 Gaussian Based Datasets 121
5.4.2 Structure Analysis of Characters 126
5.4.3 Evaluation on UCI Datasets 128
5.5 Conclusion 131
References 132
6 Fuzzy Empirical Copula for Estimating Data Dependence Structure 135
6.1 Introduction 135
6.2 Dependence Structure Estimation via Empirical Copula 137
6.2.1 Copula 138
6.2.2 Empirical Copula and Dependence Estimation 138
6.3 Fuzzy Empirical Copula 140
6.3.1 Initialisation 140
6.3.2 Approximation 141
6.3.3 Assignment 142
6.4 Experiment and Discussions 144
6.4.1 Data 144
6.4.2 Dependence Structure Estimation via Empirical Copula 146
6.4.3 Dependence Structure Estimation via Fuzzy Empirical Copula 146
6.4.4 Comparison of FLAME+ and K-Means in Fuzzy Empirical Copula 151
6.5 Conclusion 154
References 155
7 A Unified Fuzzy Framework for Human Hand Motion Recognition 158
7.1 Introduction 158
7.2 Time Clustering Recognition 160
7.2.1 Motion Model Construction 161
7.2.2 Motion Recognition 162
7.3 Fuzzy Gaussian Mixture Models Recognition 163
7.4 Recognizing with Fuzzy Empirical Copula 164
7.4.1 Re-sampling 164
7.4.2 One-to-One Correlation and Motion Template 164
7.4.3 Motion Recognition 166
7.5 Experiments 167
7.5.1 Experiment Set-Up 167
7.5.2 Experiment 1---Training with One Sample from a Single Subject 168
7.5.3 Experiment 2---Training with Six Samples from Multiple Subjects 172
7.5.4 Experiment 3---Training with Various Samples from a Single Subject 173
7.5.5 Experiment 4---Training with Various Samples from Multiple Subjects 177
7.6 Conclusion 178
References 179
8 Human Hand Motion Analysis with Multisensory Information 182
8.1 Introduction 182
8.2 Multiple-Sensor Hand Motion Capture System 185
8.2.1 System Configuration 185
8.2.2 Synchronization 186
8.2.3 Segmentation 187
8.2.4 Data Capturing 188
8.3 Correlations of Finger Trajectories, Contact Forces and EMG Signals 190
8.4 Motion Recognition via EMG Intention 192
8.4.1 Training Models via FGMMs 193
8.4.2 Recognition 194
8.4.3 Experimental Results 196
8.5 Conclusion 199
References 199
9 A Novel Approach to Extract Hand Gesture Feature in Depth Images 203
9.1 Introduction 203
9.2 Image Segmentation and De-noising Combining with Depth Information 205
9.2.1 Image Segmentation 205
9.2.2 Select the Edge Contour of Image 206
9.2.3 De-noise by Contour Length Information 206
9.3 Extract Fingertip Feature by Lasso 208
9.3.1 Earth Mover's Distance 208
9.3.2 Lasso Algorithm 210
9.4 Conclusion 213
References 214
10 Recognizing Constrained 3D Human Motion: An Inference Approach 216
10.1 Introduction 216
10.2 Human Skeletal Representation 218
10.3 The 3D Motion Recognition Framework 219
10.3.1 Fuzzy Quantile Generation 220
10.3.2 The Context-Aware Filter 225
10.3.3 Dealing with Occluded Motion 227
10.4 Experiments and Results 231
10.4.1 Experimental Setup 231
10.4.2 Evaluating Fuzzy Quantile Generation 233
10.4.3 Evaluating Occlusion Module 235
10.5 Conclusion 238
References 239
11 Study of Human Action Recognition Based on Improved Spatio-Temporal Features 242
11.1 Introduction 242
11.2 Interest Points Detection 244
11.3 Action Representation 246
11.3.1 3D SIFT Descriptor 246
11.3.2 Positional Distribution Information of Interest Points 247
11.3.3 Motion Features 248
11.4 SVM Algorithm and Results Analysis 249
11.4.1 Recognition Algorithm 249
11.4.2 Dataset 250
11.4.3 Testing Results in Portion Scenario 251
11.4.4 Testing Results in Mixed Scenarios 252
11.5 AdaBoost-SVM Algorithm and Results Analysis 255
11.5.1 AdaBoost Algorithm 255
11.5.2 Experimental Results 256
11.6 Conclusion 258
References 258
12 A View-Invariant Action Recognition Based on Multi-view Space Hidden Markov Models 260
12.1 Introduction 260
12.2 Framework of the Proposed Method 262
12.3 Feature Extraction 264
12.3.1 Interest Point Feature Extraction 264
12.3.2 Optical Flow Feature Extraction 265
12.3.3 Mixed Feature 266
12.4 Action Recognition 267
12.4.1 Sub-view Space HMM Building 267
12.4.2 Multi-view Space HMMs Probability Fusion 268
12.5 Experiments 269
12.6 Conclusion 274
References 275
Appendix A Arithmetic Operations of Chap. 2 277
Appendix B Algorithm of Chap. 4 278
Appendix C Algorithm and Proof of Chap. 5 279
Appendix D Algorithms of Chap. 6 281
Appendix E Algorithms of Chap. 9 283
Index 285

Front Matter....Pages i-xvi
Introduction....Pages 1-34
Fuzzy Qualitative Trigonometry....Pages 35-50
Fuzzy Qualitative Robot Kinematics....Pages 51-65
Fuzzy Qualitative Human Motion Analysis....Pages 67-93
Fuzzy Gaussian Mixture Models....Pages 95-121
Fuzzy Empirical Copula for Estimating Data Dependence Structure....Pages 123-145
A Unified Fuzzy Framework for Human Hand Motion Recognition....Pages 147-170
Human Hand Motion Analysis with Multisensory Information....Pages 171-191
A Novel Approach to Extract Hand Gesture Feature in Depth Images....Pages 193-205
Recognizing Constrained 3D Human Motion: An Inference Approach....Pages 207-232
Study of Human Action Recognition Based on Improved Spatio-Temporal Features....Pages 233-250
A View-Invariant Action Recognition Based on Multi-view Space Hidden Markov Models....Pages 251-267
Back Matter....Pages 269-281

These techniques are shown to effectively address the above challenges by bridging the gap between symbolic cognitive functions and numerical sensing & control tasks in intelligent systems. The book not only serves as a valuable reference source for researchers and professionals in the fields of computer vision and robotics, but will also benefit practitioners and graduates/postgraduates seeking advanced information on fuzzy techniques and their applications in motion analysis.
Erscheinungsdatum: 18.05.2017

2017-06-25