傅立成Fu, Li-Chen臺灣大學:電機工程學研究所陳羿如Chen, Yi-RuYi-RuChen2010-07-012018-07-062010-07-012018-07-062009U0001-1408200915120600http://ntur.lib.ntu.edu.tw//handle/246246/188105在這篇論文中描述了一個利用單眼相機人體上半身姿態追蹤系統。由於人體模型使用高維度的狀態空間來表示，因此利用分割取樣(partitioned sampling)降低狀態的維度，再以粒子濾波器(particle filter)來估測狀態值，其中所使用的觀測資料為影像中時間上與空間上的資訊。當進行人機互動時，靜止的單眼相機無法由影像中得到足夠的資訊，因此可以利用移動相機平台至較好的觀測位置，得到足夠的資訊來修正系統的估測值。最後，透過實驗來驗證此系統的整體效能及可靠性。This thesis presents an upper body tracking method with a monocular camera. The human model is defined in a high dimensional state space. We hereby propose a hierarchical structure model to solve the tracking problem by particle filter with partitioned sampling. The spatial and temporal information from the image is used to track the human body and estimate the human posture. When doing the human-robot interaction, a static monocular camera may not get plenty of information from the 2D images, so we must move the camera platform to a better position for acquiring more enriched image information. The proposed upper body tracking technique will then self-adjust to estimate the human posture during the camera movement. To validate the effectiveness of the proposed tracking approach, extensive experiments have been performed, of which the result appear to be quite promising.摘要 IBSTRACT IIONTENTS IIIIST OF FIGURES VIST OF TABLES VIIHAPTER 1 INTRODUCTION 1.1 MOTIVATION 2.2 RELATED WORKS 3.3 CONTRIBUTION 5.4 THESIS ORGANIZATION 6HAPTER 2 PRELIMINARIES 7.1 BAYESIAN FILTER 7.2 PARTICLE FILTER 12.2.1 Monte Carlo Integration 13.2.2 Sequential Importance Sampling (SIS) Particle Filter 14.2.3 Resampling and Degeneracy Problem 17.2.4 Sampling Importance Resampling (SIR) Particle Filter 20.2.5 Impoverishment Phenomenon 21.3 PARTITIONED SAMPLING 22.4 OPTICAL FLOW 26HAPTER 3 UPPER BODY TRACKING 29.1 HUMAN MODEL DEFINITION 29.2 POSTURE INITIALIZATION 36.3 PARTITIONED SAMPLING FOR HUMAN MODEL 38.4 SIR PARTICLE FILTER FOR FACE TRACKER 39.5 MULTIPLE IMPORTANCE SAMPLING (MIS) 40.5.1 The Balance Heuristic 41.6 MIS PARTICLE FILTER FOR ARM TRACKER 44.6.1 Latest Posterior 46.6.2 Inverse Kinematics 47.6.3 Line Detector 49.7 UPPER BODY TRACKING SYSTEM 51HAPTER 4 IMPLEMENTATION 53.1 LIKELIHOOD FUNCTION FOR PARTICLE FILTER 53.1.1 Color Histogram 54.1.2 Enhanced Edge Contour 56.1.3 Angle on Image Plane 61.1.4 Overall Likelihood 62.2 HUMAN ROBOT INTERACTION 63.3 MOTION STRATEGY 65HAPTER 5 EXPERIMENTAL RESULT 67.1 ENVIRONMENT DESCRIPTION 67.2 RESULTS OF UPPER BODY TRACKING 68.3 RESULTS OF MOTION CAMERA 71.4 RESULTS OF HUMAN ROBOT INTERACTION 73.5 DISCUSSIONS 75HAPTER 6 CONCLUSION AND FUTURE WORK 77.1 CONCLUSION 77.2 FUTURE WORK 78EFERENCE 791660630 bytesapplication/pdfen-US粒子濾波器影像追蹤上半身追蹤particle filtervisual trackingupper body trackingthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/188105/1/ntu-98-R96921005-1.pdf