Journal of the Institute of Electronics Engineers of Korea SP (대한전자공학회논문지SP)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Non-Prior Training Active Feature Model-Based Object Tracking for Real-Time Surveillance Systems

실시간 감시 시스템을 위한 사전 무학습 능동 특징점 모델 기반 객체 추적

  • Published : 2004.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper we propose a feature point tracking algorithm using optical flow under non-prior taming active feature model (NPT-AFM). The proposed algorithm mainly focuses on analysis non-rigid objects[1], and provides real-time, robust tracking by NPT-AFM. NPT-AFM algorithm can be divided into two steps: (i) localization of an object-of-interest and (ii) prediction and correction of the object position by utilizing the inter-frame information. The localization step was realized by using a modified Shi-Tomasi's feature tracking algoriam[2] after motion-based segmentation. In the prediction-correction step, given feature points are continuously tracked by using optical flow method[3] and if a feature point cannot be properly tracked, temporal and spatial prediction schemes can be employed for that point until it becomes uncovered again. Feature points inside an object are estimated instead of its shape boundary, and are updated an element of the training set for AFH Experimental results, show that the proposed NPT-AFM-based algerian can robustly track non-rigid objects in real-time.

본 논문에서는 사전학습이 필요 없는 능동 특징점 모델(non-prior training active feature model; NPT AFM) 기반에서 광류(optical flow)를 이용한 객체추적 기술을 제안한다. 제안한 알고리듬은 비정형 객체에 대한 분석[1]에 초점을 두고 있으며, 실시간에서 NPT-AFM을 사용한 강건한 추적을 가능하게 한다. NPT-AFM 알고리듬은 관심 객체의 위치를 파악하는 과정 (localization)과 이전 프레임 정보와 현재 프레임 정보를 이용하여, 객체의 위치를 예측(prediction), 보정(correction)하는 과정으로 나눌 수 있다 위치 파악 과정에서는 움직임 분할(motion segmentation)을 수행한 후 개선된 Shi-Tomasi의 특징점 추적 알고리듬[2]을 사용 하였다. 예측 및 보정 과정에서는 광류 정보를 사용하여 특징점을 추적하고[3] 만약, 특징점이 적절히 추적 되지 않거나 추적에 실패하면 특징점들의 시간(temporal), 공간(spatial)적 정보를 이용하여 예측, 보정하게 된다. 객체의 형태 (shape)대신 특징점을 사용하였으며, 객체를 추적하는 과정에서 특징점들은 능동 특징점 모델(active feature model; AFM)을 위한 학습 집합(training sets)의 요소로 갱신된다. 실험결과, 제안한 NPT-AF% 기반 추적 알고리듬은 실시간에서 비정형 객체를 추적하는데 강건함을 보석준다.

Keywords

References

  1. C. Eroglu, M. Tekalp, and B. Sankur, 'Non-Rigid Object Tracking Using Performance Evaluation Measures as Feedback,' Proc. IEEE Int. Conf. Computer Vision and Pattern Recognition, pp. 323-330, 2001 https://doi.org/10.1109/CVPR.2001.990978
  2. J. Shi and C. Tomasi, 'Good features to track,' in Proc. IEEE int. Conf. Computer Vision and Pattern Recognition, pp. 593-600, 1994 https://doi.org/10.1109/CVPR.1994.323794
  3. D. Bruce, Lucas and Takeo Kanade, 'An Iterative Image Registration Technique with an Application to Stereo Vision,' In Proc. DARPA image Understanding Work-shop, pp. 121-130, 1981
  4. S. McKenna, Y. Raja, and S. Gong, 'Tracking Contour Objects Using Adaptive Mixture Models,' Image and Vision Computing, pp. 225-231, 1999 https://doi.org/10.1016/S0262-8856(98)00104-8
  5. R. Plankers and P. Fua, 'Tracking and Modeling People in Video Sequences,' Computer Vision and Image Understanding, pp. 285-302, 2001 https://doi.org/10.1006/cviu.2000.0891
  6. D. Comaniciu, V. Ramesh and, P. Meer, 'Kernel-based object tracking,' IEEE Trans. On Pattern Analysis and Machine Intelligence, vol. 25, No. 5, pp. 564-577, May 2003 https://doi.org/10.1109/TPAMI.2003.1195991
  7. I. Haritaolu, D. Harwood and L. S. Davis, 'W4: real-time surveillance of people and their activities,' IEEE Trans. on PAMI, 22(8): 809-830, 2000 https://doi.org/10.1109/34.868683
  8. C. Wren, A. Azarbayejani, T. Darrell and A. Pentland, 'Pfinder: Real-time tracking of the human body,' IEEE trans. on PAMI, 19(7):780-785, 1997 https://doi.org/10.1109/34.598236
  9. D. Comaniciu, V. Ramesh and P. Meer, 'Real-time tracking of non-rigid objects using mean shift,' in Proc. IEEE int. Conf. Computer Visoin and Pattern Recognition, vol. 2, Hilton Head Island, SC, pp. 142-149, June 2000 https://doi.org/10.1109/CVPR.2000.854761
  10. A. M. Baumberg, 'Learning deformable models for tracking human motion,' Ph.D. dissertation, School of Comput. Studies, Univ. of Leeds, U.K., Oct 1995
  11. T. J. Cootes, C. J. Taylor, D. H. Cooper, and J. Gragam, 'Training models of shape form sets of examples,' In British Machine Vision Conference, pp. 9-18, September 1992
  12. M. Isard and A. Blake, 'Condensation conditional density propagation for visual tracking,' Int. J. Comput. Vis, vol. 29, no. 1, pp. 5-28, 1998 https://doi.org/10.1023/A:1008078328650
  13. A. Koschan, S. Kang, J. Paik, B. Abidi, and M. Abidi, 'Color active shape models for tracking non-rigid objects,' Pattern Recognition Letters, vol. 24, no. 11, pp. 1751-1765, July 2003 https://doi.org/10.1016/S0167-8655(02)00330-6
  14. C. Goodall. Procrustes methods in the statistical analysis of shape. Journal of the Royal Statistical Society B, 53(2): 285-339, 1991
  15. T. F. Cootes, C. J. Taylor, and A. Lanitis, 'Active shape models: evaluation of a multi-resolution method for improving image search,' Proc. British Machine Vision Conference, pp. 327-336, 1994