The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

Bilateral Filtering-based Mean-Shift for Robust Face Tracking

양방향 필터 기반 Mean-Shift 기법을 이용한 강인한 얼굴추적

  • 최완용 (광운대학교 제어계측공학과) ;
  • 이윤형 ((주)만도 중앙연구소) ;
  • 정문호 (광운대학교 로봇학부)
  • Received : 2013.07.25
  • Accepted : 2013.09.23
  • Published : 2013.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

The mean shift algorithm has achieved considerable success in object tracking due to its simplicity and robustness. It finds local minima of a similarity measure between the color histograms or kernel density estimates of the target and candidate image. However, it is sensitive to the noises due to objects or background having similar color distributions. In addition, occlusion by another object often causes a face region to change in size and position although a face region is a critical clue to perform face recognition or compute face orientation. We assume that depth and color are effective to separate a face from a background and a face from objects, respectively. From the assumption we devised a bilateral filter using color and depth and incorporate it into the mean-shift algorithm. We demonstrated the proposed method by some experiments.

Mean-Shift 알고리즘은 목표모델과 후보영상 사이의 컬러분포의 유사도를 이용하는 국부적 탐색기법으로서, 그 기법의 단순성 및 안정성 면에서 뛰어나 얼굴추적에 많이 이용되고 있다. 그러나 컬러분포를 이용한 얼굴추적은 목표모델과 유사한 컬러분포를 갖는 객체나 배경의 영향에 취약하다. 또한 얼굴 추적에서 결정되는 얼굴영역은 얼굴인식 혹은 얼굴방향 등을 계산할 때 중요한 단서가 되는데, 완전히 다른 컬러분포를 갖는 객체의 가려짐으로 얼굴영역의 크기나 위치가 변동될 위험이 있다. 대체로 거리정보는 얼굴과 배경의 구분에 효율적이고 컬러정보는 객체 구분에 유리하다는 가정으로부터, 본 논문에서는 이러한 문제를 해결하기 위해 거리 정보와 컬러 정보를 함께 이용하는 양방향 필터를 고안하고, 이것을 Mean-Shift 알고리즘에 활용하였다. 일련의 실험을 통해 성공적인 실험결과를 얻었다.

Keywords

References

  1. Y. Cui, S. Samarasekera, Q. Huang, M Greienhagen, "Indoor Monitoring Via the Collaboration Between a Peripheral Senson and a Foveal Sensor", IEEE Workshop on Visual Surveillance, Bombay, India, pp. 2-9, 1998.
  2. Jang-sik Park, Hyun-tae Kim, Yun-sik Yu, "Video Based Fire Detection Algorithm using Gaussian Mixture Model", The Korea Institute of Electronic Communication Science, Vol. 6, No. 2, pp. 206-211, 2011.
  3. Ik-Soon Kim, Hyun-shik Shin, "A Study on Development of Intelligent CCTV Security System based on BIM", The Korea Institute of Electronic Communication Science, Vol. 6, No. 5, pp. 789-795, 2011.
  4. Ik-soon Kim, Jae-duck Yoo, Bae-hun Kim, "A Monitoring Way and Installation of Monitoring System using Intelligent CCTV under the u-City Environment", The Korea Institute of Electronic Communication Science, Vol. 3, No. 4, pp. 295-303, 2011.
  5. G.R. Bradski, "Computer Vision Face Tracking as a Component of a Perceptual User Interface," IEEE Work. on Applic. Comp. Vis., Princeton, pp. 214-219, 1998.
  6. R. Chellappa, S. Zhou, and B. Li. Bayesian methods for face recognition from video. In Int. Conf. on Acoustics Speech and Signal Processing, Orlando, Florida, 2002.
  7. C. Wren, A. Azarbayejani, T. Darrell, A. Pentland, "Pfinder: Real-Time Tracking of the Human Body," IEEE Trans. Pattern Analysis Machine Intell., Vol. 19, pp. 780-785, 1997. https://doi.org/10.1109/34.598236
  8. A. Eleftheriadis, A. Jacquin, "Automatic Face Location Detection and Tracking for Model- Assisted Coding of Video Teleconference Sequences at Low Bit Rates," Signal Processing - Image Communication, Vol. 7, No. 3, pp. 231-248, 1995. https://doi.org/10.1016/0923-5965(95)00028-U
  9. D. Comaniciu, V. Ramesh, and P. Meer. Kernel-based object tracking. IEEE Trans. Pattern Anal. Mach. Intell., Vol. 25, No. 5, pp. 564-577, 2003. https://doi.org/10.1109/TPAMI.2003.1195991
  10. Paul Viola, M. Jones, Rapid Object Detection using a Boosted Cascade of Simple, Proceedings of CVPR2001, Vol. I, pp. 511-518, 2001.