人體運動參數的偵測由於它廣泛的應用面，最近幾年在影像處理方面被密切重視。而這些人體運動參數偵測的研究通常會遇到兩個問題，第一個問題是人體運動參數的高維度，第二個問題是當遮蔽發生而導致資訊減少時的運動參數偵測。為了解決這兩個問題，我們在這裡提出一個基於影像處理的方法，這個方法是結合一個從Particle Filter改良而來的Annealed Particle Filter (APF)[6]以及一個事先訓練好的correlation map和temporal constraint 去做人體走路參數的偵測。
在本篇論文中，我們會做各種不同拍攝角度的走路參數偵測，首先我們會使用OpenGL建構出3D模型，然後將人體模型分成10個部分，並由12維度的走路參數來表示各種的走路姿態，我們會分別使用形狀和顏色的資訊來作為我們做走路參數偵測的依據，接下來我們會將APF結合correlation map和temporal constraint做各個走路參數的偵測，接著就將我們偵測到的結果使用OpenGL繪製出來
並且，我們提供了一個有效的人體運動參數偵測可運行於室內以及戶外的環境下。戶外環境相較於室內環境最大的問題就是影子所造成的干擾，所以我們會在將影像轉換成在HSV的維度下，然後對影子做處理。由於我們加入了correlation map和temporal constraint的觀念，所以相較於傳統的APF[6]我們可以大幅的縮短運算時間，並且有效的增加運動參數估測的準確度。另一方面，當身體各個部分發生互相遮蔽的時候，相較於傳統的APF，使用我們的方法，在實驗結果上也可以發現明顯的改善。

Markerless human body part tracking and pose estimation have recently attracted intensive attention because of their wide applications. The vision-based approaches to solve the problems of motion parameters capturing always meet two challenges. 1) how to solve the parameter estimation problem in high-dimensional space, and 2) how to deal with the missing observation information due to occlusion. To solve the two problems, we proposed a vision-based method combining the Annealed Particle Filter (APF) [6] with a pre-trained correlation map and temporal constraint.
This paper proposes a system for capturing motion parameters of walking human object in indoors and outdoors. To solve the problem with shadow when we track the walking people in outdoors, we use the HSV model to remove the shadow. Compare to the traditional APF [6], our method needs less operation time and has more accurate result. Because of the pre-trained correlation map and temporal constraint, our method also has better performance than the tradition APF when self-occlusion occurs.

Chapter 1 Itroduction……………………………………1
1.1 Motivation………………………………………1
1.2 Related Works..…………………………………2
1.3 System Overview…………………………………5
Chapter 2 3-D Human Model…………………………………7
2.1 Model Constraint…………………………………9
2.2 The Ratio of limbs……………………………11
Chapter 3 Annealed Particle Filter………………………13
3.1 Particle Filter……………………………………13
3.2 Annealed Particle Filter…………………………16
3.3 Compare Annealed Particle Filter with Particle Filter……………………………19
Chapter 4 Walking Motion Parameters Capturing…………23
4.1 Foreground Segmentation……………………………24
4.1.1 Simple Background Subtraction………………25
4.1.2 Foreground Size Filter…………………………27
4.1.3 Shadow Removed in HSV Domain…………………28
4.2 Color Histogram Information………………………29
4.3 The Weighting Function by Observation…………30
4.4 The Joint Angle Spatial Correlation……………31
4.5 The Temporal Constraint for the Same Body Part…35
Chapter 5 Experimental Results……………………………38
Chapter 6 Conclusions and Future Work……………………52
References………………………………………………………53

[1] C. Chen and G. Fan” Combining Spatial and Temporal Priors for Articulated Human Tracking with Online Learning” IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops 2009.
[2] G.C. Goodwin and J.C ” State and Parameter Estimation for Linear and Nonlinear Systems” Proc of the 7th International Conf. On Control Automation, Robotics and Vision. 2002.
[3] H. Su and F.G. Huang “Human Gait Recognition Based On Motion Analysis ” Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on pages 4464 - 4468 Vol. 7
[4] J. Gall. “Generalised Annealed Particle Filter -Mathematical Framework, Algorithms and Applications”, Diploma thesis of University of Mannheim, Department of Mathematics and Computer Science, 2005
[5] J. Darby, B. Li and N. Costen “Tracking a Walking Person using Activity-Guided Annealed Particle Filtering” Automatic Face & Gesture Recognition, 2008. FG '08. 8th IEEE International Conference on
[6] J. Deutscher, Andrew Blake, and Ian Reid” Articulated Body Motion Capture by Annealed Particle Filtering” CVPR 2000
[7] L. Raskin, E. Rivlin, M. Rudzsky ”Dimensionality reduction for articulated body tracking” 3DTV Conference, 7-9 May 2007 in Kos Island
[8] L.Q. Li, H.B. Ji, J.H. Luo ”The Iterated Extended Kalman Particle Filter”, Communications and Information Technology, 2005. ISCIT 2005. IEEE International Symposium on
[9] P. Peursum. “On the Behaviour of Body Tracking with the Annealed Particle Filter in Realistic Conditions” TECHNICAL REPORT of the Dept of Computing, Curtin University of Technology G.2006
[10] P. Peursum, S. Venkatesh and G. West “Tracking-as-Recognition for Articulated Full-Body Human Motion Analysis” CVPR 2007
[11] R. Rosipal and N. Kramer. “Overview and recent advances in Partial Least Squares.” Subspace, Latent Structure and Feature Selection, Statistical and Optimization, Perspectives Workshop, SLSFS 2005.
[12] T. X. Han Huazhong Ning Thomas S. Huang “ Efficient Nonparametric Belief with Application to Articulated Body Tracking” CVPR 2006
[13] W. Noorshahida Mohd Isa1, Rubita Sudirman2, Sheikh Hussain Sh-Salleh “Angular Features Analysis for Gait Recognition” Computers, Communications, & Signal Processing with Special Track on Biomedical Engineering, 2005. CCSP 2005. 1st International Conference on 2005.
[14] W. Guo, C. Han, M. Lei “Improved Unscented Particle Filter for Nonlinear Bayesian Estimation” Information Fusion, 2007 10th International Conference on,2007
[15] X. Xu, B. Li” Learning Motion Correlation for Tracking Articulated Human Body with a Rao-Blackwellised Particle Filter” Computer Vision, 2007. ICCV 2007. IEEE 11th International Conference on 2007.