The Journal of Korean Institute of Communications and Information Sciences (한국통신학회논문지)

The Korean Institute of Commucations and Information Sciences (한국통신학회)
권호 표지
DOI QR코드

DOI QR Code

Face Tracking Combining Active Contour Model and Color-Based Particle Filter

능동적 윤곽 모델과 색상 기반 파티클 필터를 결합한 얼굴 추적

  • Received : 2015.07.08
  • Accepted : 2015.10.13
  • Published : 2015.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

We propose a robust tracking method that combines the merits of ACM(active contour model) and the color-based PF(particle filter), effectively. In the proposed method, PF and ACM track the color distribution and the contour of the target, respectively, and Decision part merges the estimate results from the two trackers to determine the position and scale of the target and to update the target model. By controlling the internal energy of ACM based on the estimate of the position and scale from PF tracker, we can prevent the snake pointers from falsely converging to the background clutters. We appled the proposed method to track the head of person in video and have conducted computer experiments to analyze the errors of the estimated position and scale.

본 논문은 ACM(active contour model)과 색상기반 PF(particle filter)의 장점을 결합하여 크기와 색상이 변화하는 객체에 대해 강인한 추적이 가능한 방법을 제안한다. 제안하는 방법은 색상기반의 PF 추적기, 윤곽선을 추적하는 ACM 추적기, 그리고 두 추적기의 추정 정보를 결합하여 최종적인 객체의 위치와 스케일을 결정하고 또 참조 모델의 업데이트 여부를 결정하는 Decision 부로 이루어진다. PF 추적기는 객체의 형태변화와 모션블러에 강인하지만 위치와 스케일의 정확도가 떨어지고, ACM 추적기는 배경 클러터가 없는 경우에는 객체의 윤곽을 정확하게 추출하지만 복잡한 배경에서는 추적에 실패하는 문제가 있다. 본 논문에서는 색상 PF 추적기가 추정한 객체 위치와 스케일 정보를 이용하여 ACM의 내부 에너지를 제어함으로써 ACM의 스네이크 포인터가 객체가 아닌 배경 클러터로 수렴되는 것을 방지하여 정확히 객체의 윤곽을 추적할 수 있도록 하였다. 사람의 머리 윤곽선을 포함한 얼굴 추적에 제안된 알고리즘을 적용하고 추정 위치와 스케일 오차를 분석하여 성능을 분석하였으며 제안된 방식이 기존 기법들보다 추적 성능이 우수함을 보였다.

Keywords

References

  1. J.-Y. Nam and J. Y. Kwak, "Object tracking using particle filters in moving camera," J. KICS, vol. 37A, no. 05, pp. 375-387, May 2012.
  2. D.-H. Kim and G.-H. Kim, "Detection of objects temporally stop moving with spatio-temporal segmentation," J. KICS, vol. 40, no. 01, pp. 142-151, Jan. 2015. https://doi.org/10.7840/kics.2015.40.1.142
  3. J.-M. Choi, H. Song, S. H. Park, and C.-D. Lee, "Implementation of driver fatigue monitoring system," J. KICS, vol. 37C, no. 08, pp. 711-720, Aug. 2012.
  4. J. T. Ryu, J. M. Yang, Y. S. Choi, and S. H. Park, "Improving the processing speed and robustness of face detection for a psychological robot application," J. Korea Ind. Inf. Syst. Research, vol. 20, no. 2, pp. 57-63, Apr. 2015. https://doi.org/10.9723/JKSIIS.2015.20.2.057
  5. K. Jeong, J. Choi, and G.-J. Jang, "Facial expression recognition using face alignment and AdaBoost," J. Inst. Electron. and Inf. Eng., vol. 51, no. 11, pp. 193-201, Nov. 2014. https://doi.org/10.5573/IEIE.2014.51.11.193
  6. Y.-J. Bae, H.-J. Choi, Y.-H. Seo, and D.-W. Kim, "A fast and accurate face detection and tracking method by using depth information," J. KICS, vol. 37A, no. 07, pp. 586-599, Jul. 2012.
  7. D. Cho, S. Lee, and I. H. Suh, "Facial feature tracking using adaptive particle filter and active appearance model," J. Korea Robotics Soc., vol. 8, no. 2, pp. 104-115, Jun. 2013. https://doi.org/10.7746/jkros.2013.8.2.104
  8. W. Kim and J. Chun, "A hybrid approach of efficient facial feature detection and tracking for real-time face direction estimation," J. Internet Computing and Services (JICS), vol. 14, no. 6, pp. 117-124, Dec. 2013. https://doi.org/10.7472/jksii.2013.14.6.117
  9. D. Comaniciu, V. Ramesh, and P. Meer, "Real-time tracking of non-rigid objects using mean shift," in Proc. Computer Vision and Pattern Recognition, pp. 142-149, 2000.
  10. W.-Y. Choi, Y.-H. Lee, and M.-H. Jeong, "Bilateral filtering-based mean-shift for robust face tracking," J. KIECS, vol. 8, no. 9, pp. 1319-1324, 2013.
  11. K. Nummiaro, E. Koller-Meier, and L. Van Gool, "An adaptive color-based particle filter," Image and Vision Computing, vol. 21, no. 1, pp. 99-110, Jan. 2003. https://doi.org/10.1016/S0262-8856(02)00129-4
  12. M. Kass, "Snake: active contour model," Int. J. Computer Vision, vol. 1, no. 4, pp. 321-331, 1988. https://doi.org/10.1007/BF00133570
  13. J. H. Lee, H. G. Oh, and H. Hong, "Active contour model for object tracking with large motion displacement," in Proc. 33rd Korea Inf. Sci. Soc. Conf., vol. 33, no. 2(B), pp. 464-469, Sejong Univ., Seoul, Korea, Oct. 2006.
  14. G. D. Giannoglou, et al., "A novel active contour model for fully automated segmentation of intravascular ultrasound images," Computers in Biol. Med., vol. 37, no. 9, pp. 1292-1302, Sept. 2007. https://doi.org/10.1016/j.compbiomed.2006.12.003
  15. J.-Y. Kim and J.-K. Jeong, "Object contour tracking using an improved snake algorithm," J. IEEK - Signal Processing, vol. 48-SP, no. 6, pp. 105-114, Nov. 2011.