這篇論文的目標，在於追蹤輪椅，並提出在如何由單一視角的影像序列中，快速且正確的得到三度空間的輪椅位置與方向的方法。在這篇論文中，三度空間的定位，是利用橢圓與圓的投影幾何關係求得。我們可以從輪椅輪子在二度空間中投影出的圓狀軌跡，得到其三度空間的相關資訊。我們需要橢圓方程式來達到我們的目地，在此我們使用隨機哈夫轉換和最小平方誤差逼近法來得到橢圓方程式當作3D定位演算法的輸入。由此我們可以精確的得到輪椅於三度空間中的位置與方向。接著我們使用延伸卡爾曼濾波器並以加入方向考量的運動方程式來做為追蹤及預測的輪椅運動的工具。實驗結果顯示，我們的方法能有相當理想的結果。

This paper aims at tracking a wheelchair from monocular image sequences and presents an efficient and robust approach to retrieve three-dimensional (3D) wheelchair pose. In this paper, 3D localization is achieved by ellipse-circle geometry, which only requires ellipse equation extraction instead of mapping features correspondences between 2D image features and the 3D model. The ellipse equation is obtained by RHT and least square ellipse fitting to provide input for the 3D localization. Based on the geometry, we are able to provide accurate result for wheelchair pose. Then the EKF is utilized to track and predict the wheelchair motion with orientation-included 3D dynamic model. Experiment results show that the algorithm obtains desirable performance.

[1] J. Philip “An algorithm for determining the position of a circle in 3d from its perspective 2D projection,” TRITA-MAT-1997-MA-14, 1997.
[2] A. Myles, N,D.V Lobo, Mubarak Shah, “Wheelchair Detection in a Calibrated Environment” The 5th Asian Conference on Computer Vision, ,Melbourne, Australia ,2002.
[3] D. Chen, J. Yang, H. D. Wactlar, “Towards automatic analysis of social interaction patterns in a nursing home environment from video.” International Multimedia Conference Proceedings of the 6th ACM SIGMM international workshop on Multimedia information retrieval. P.283-290 ,2004.
[4] J.J. Wu, R. E. Rink, T. M. Caelli, and V. G. Gourishankar. “Recovery of the 3-D Location and Motion of a Rigid Object through Camera Image”. International Journal of Computer Vision, vol. 3, pp.373-394, 1988
[5] R. Hartley and A. Zisserman, “Multiple View Geometry in Computer Vision, 2nd ed.” Cambridge University Press, 2003, pp.153-157.
[6] K. Kanatani, “3D Recovery from Angle Clues” Proc. 5th Scandinavian Conference on Image Analysis, Stockholm. pp. 167-179,June 1984.
[7] T. Shakunaga and H. Kaneko, “Perspective Angle Transform and Its Application to 3-D Configuration Recovery”, Proc. IEEE Conference on computer Vision and Pattern Recognition, Miami Beach ,Fl., pp.594-601 ,June 1986.
[8] Z. Zhang , “A Flexible New Technique for Camera Calibration”, Technical Report MSR-TR-98-71, Microsoft Research. 1998
[9] J. Wang, W. J. Wilson, “3D Relative Position and Orientation Estimation Using Kalman Filter For Robot Control”, Proceedings of the 1992 IEEE International Conference on Robotics and Automation, Nice, France, May 1992
[10] J. S.C Yuan, “A general photogrammetric method for determining object position and orientation.” IEEE Trans. Robotics Auto., p129, 1989.
[11] R. E. Kalman, “A New Approach to Linear Filtering and Prediction Problems”, Transaction of the ASME – Journal of Basic Engineering pp.35-45, March 1960.
[12] J. Canny, “A Computational Approach to Edge Detection”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 8, No. 6, Nov 1986.
[13] K. C. Chang, H. J Lee, and C. G. Chung, “Adaptively estimation motion parameters for tracking a maneuvering target using images”, Inter. Robotics and Automation, p 43, 1989.
[14] R. A. McLaughlin , “Randomized Hough Transform: Improved Ellipse Detection with Comparison”, Pattern Recognition Letters, 1998
[15] G. Welch, G. Bishop, “An Introduction to Kalman Filter”, ACM SIGGRAPH 2001 Course Notes, 2001.
[16] H. K. Yuen, J. Illingworth, and J. Kittler, “Detecting partially occluded ellipse using Hough Transform”, Image and Vision Computing, vol. 7, no. 1, p 31-37, 1989.
[17] S. J. Julier, and J. K. Uhlmann, “A new extension of the Kalman Filter to Nonlinear Systems”, SPIE AeroSense Symposium, April 21–24, 1997.
[18] Zs. Lendek, R. Babuska, B. De Schutter, “Nonlinear State Estimation for Control: Comparison of UKF and PF”. Abstract for a talk within The 25th Benelux Meeting on Systems and Control, Heeze, The Netherlands, 13-15 March 2006
[19] N. Werghi, C. Doignon, and G. Abba, “Ellipse fitting and three-dimensional localization of object based on ellipse features”, International Conference on Image Processing, 1996.
[20] Zs. Lendek, R. Babu‧ska and B. De Schutter, “State Estimation under Uncertainty : A survey”, Technical Report 06-004, February 2006
[21] J. D. Yoder, E. Baumartner, S. B. Skaar, “Reference Path Description for an Autonomous Powered Wheelchair”, IEEE Transactions on Rehabilitation Engineering, 1996.
[22] M. S. Grewal, A. P. Andrews, “Kalman Filtering, Theory and Practice Using MATLAB, Second Edition”, Wiley Inter-science, 2001.
[23] J. Lou, T. Tan, W. Hu, H. Yang, S. J. Maybank, “3-D Model-Based Vehicle Tracking”, IEEE Transactions on Image Processing, vol. 14, no. 10, October 2005.
[24] T. N. Tan, G. D. Sullivan, and K. D. Baker, “Model-based localization and recognition of road vehicles”, International Journal on Computer Vision, vol. 27, no.1, pp.5-25, 1998.
[25] P. Beardsley, “Wheelchair detection using stereo vision”, MERL TR2007-044, January, 2007
[26] L. M. Bergasa, M. Mazo, A. Gardel, J. C. Garcia, A. Ortuo, A. E. Mendez, “Guidance of a wheelchair for handicapped people by face tracking”, Proceedings. ETFA, October, 1999.
[27] C. Cauchois, F. de Chaumont, B. Marhic, L. Delahoche, M. Delafosse., “Robotic assistance: an automatic wheelchair tracking and following functionality by omnidirectional vision”, International Conference on IEEE/RSJ, August 2005.
[28] D. Roller, K. Daniilidis, and H. N. Nagel., “Model-based object tracking in monocular image sequence of road traffic scenes”, International Journal of Computer Vision, vol. 10. no. 3, June 1993.
[29] H.T. Chen, H. H. Lin, T. L. Liu, “Multi-object tracking using dynamical graph matching”, IEEE Computer Society Conference on Computer Vision Pattern Recognition, vol.2 , pp. 210-217,2001.