Publications
Detailed Information
Gait Pattern Prediction and Gait-Based Human Identification Using Machine Learning Methods : 기계학습 기반 기법을 이용한 보행 동작 예측과 보행 기반 개인 식별
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 박종우 | - |
dc.contributor.author | 홍지수 | - |
dc.date.accessioned | 2018-11-12T00:56:52Z | - |
dc.date.available | 2018-11-12T00:56:52Z | - |
dc.date.issued | 2018-08 | - |
dc.identifier.other | 000000153599 | - |
dc.identifier.uri | https://hdl.handle.net/10371/143127 | - |
dc.description | 학위논문 (박사)-- 서울대학교 대학원 : 공과대학 기계항공공학부, 2018. 8. 박종우. | - |
dc.description.abstract | This thesis is concerned with human gait motion learning and prediction,
focusing on differences in gait motion between individuals to address the problems of personalized gait pattern prediction and gait based biometric identification. We first present an algorithm that predicts individual gait patterns from a human's anthropometric characteristics, e.g., age, height, weight, gender, and other related anthropometric data. We adopt the Gaussian process dynamical model~(GPDM) framework to address the high dimensionality of human gait motions, and to learn a common stochastic dynamics model for gaits. We also utilize Gaussian process regression~(GPR) to learn a mapping from the space of body features to the motion parameters in the GPDM framework. Using our framework, an entire cycle of individualized gait motions at arbitrary walking speeds can be predicted from body feature data. We also propose a gait-based identification algorithm based on deep neural networks. To allow for new subjects whose information is not included in the original database, we develop a novel autoencoder architecture that is designed to extract features efficiently from the measured gait motions. We use the reconstruction error of the trained autoencoder as a similarity measure between the learned motion and the input, for which user-specified thresholds can be used for the identification. Our proposed method is able to identify both subjects in the database as well as unknown subjects, only using the observed gait motion. Furthermore, it is possible to update the database incrementally for newly given gait motions. Finally, we propose an algorithm used to predict a subject's age, height, weight, and gender from the subject's gait motion. We construct a neural network architecture for this purpose by modifying the previous autoencoder structure used for the identification, and devise an efficient learning strategy utilizing transfer learning to reduce the prediction error. To validate proposed algorithms, we collect gait motions and anthropometric data from more than a hundred subjects. For our gait motion prediction algorithm, an individualized gait pattern is generated for observed anthropometric data and at arbitrary walking speeds. The algorithm also shows 30% less gait pattern prediction errors compared to a frame-by-frame statistical regression method. In the case of gait identification, our algorithm shows identification accuracy of 98.5-100%, outperforming the results reported in recent studies. | - |
dc.description.tableofcontents | 1 Introduction 1
1.1 Research Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.2 Contributions of This Thesis . . . . . . . . . . . . . . . . . . . . . . 7 1.2.1 Gait Pattern Prediction from Individual Physical Features . 7 1.2.2 Gait Pattern-Based Human Identification . . . . . . . . . . . 9 1.2.3 Gait Pattern-Based Physical Feature and Gender Prediction 10 1.3 Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2 Preliminaries 13 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 2.2 Human Motion Modeling with Nonlinear Dimensionality Reduction 13 2.2.1 Principal Component Analysis . . . . . . . . . . . . . . . . . 14 2.2.2 Geometrically-Inspired Algorithms . . . . . . . . . . . . . . . 16 2.2.3 Gaussian Process-Based Models . . . . . . . . . . . . . . . . 17 2.3 Deep Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.3.1 Feedforward Network . . . . . . . . . . . . . . . . . . . . . . 29 2.3.2 Autoencoder . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.3.3 Recurrent Neural Network . . . . . . . . . . . . . . . . . . . 31 2.3.4 Optional Elements . . . . . . . . . . . . . . . . . . . . . . . . 34 3 Gait Pattern Prediction from Individual Physical Features 39 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.2.1 Acquisition and Preprocessing of Human Gait Motion . . . 43 3.2.2 Gaussian Process Learning of Gait Motion . . . . . . . . . . 45 3.2.3 Generation of Individualized Gait Motion . . . . . . . . . . . 50 3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.3.1 Determination of the Dimension of the Latent Space . . . . 52 3.3.2 Gait Patterns at Intermediate Walking Speed . . . . . . . . 53 3.3.3 Prediction of the Initial Poses . . . . . . . . . . . . . . . . . 55 3.3.4 Generation of the Individualized Gait Pattern . . . . . . . . 56 3.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 4 Gait Pattern-Based Human Identification 63 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 4.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.2.1 Gait Identification for Known Classes . . . . . . . . . . . . . 66 4.2.2 Gait Identification Including Unknown Classes . . . . . . . . 68 4.3 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.3.1 Gait Identification with Known Subjects . . . . . . . . . . . 71 4.3.2 Gait Identification with Known and Unknown Subjects . . . 75 4.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 5 Gait Pattern-Based Physical Feature and Gender Prediction 81 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 5.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 5.2.1 Transfer Learning Scheme for Feature Estimation . . . . . . 83 5.2.2 Experiments and Results . . . . . . . . . . . . . . . . . . . . 85 5.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 6 Conclusion 93 Bibliography 97 Abstract 115 | - |
dc.language.iso | en | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject.ddc | 621 | - |
dc.title | Gait Pattern Prediction and Gait-Based Human Identification Using Machine Learning Methods | - |
dc.title.alternative | 기계학습 기반 기법을 이용한 보행 동작 예측과 보행 기반 개인 식별 | - |
dc.type | Thesis | - |
dc.contributor.AlternativeAuthor | Hong Jisoo | - |
dc.description.degree | Doctor | - |
dc.contributor.affiliation | 공과대학 기계항공공학부 | - |
dc.date.awarded | 2018-08 | - |
- Appears in Collections:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.