DOI QR코드

DOI QR Code

정밀하지 않은 깊이정보와 2D움직임 정보를 이용한 사용자 검출과 주요 신체부위 추정

User Detection and Main Body Parts Estimation using Inaccurate Depth Information and 2D Motion Information

  • 이재원 (전남대학교 전자컴퓨터공학과) ;
  • 홍성훈 (전남대학교 전자컴퓨터공학부, 정보통신연구소)
  • Lee, Jae-Won (Division of electronic computer Engineering, Chonnam National Univ.) ;
  • Hong, Sung-Hoon (School of Electronic & Computer Engineering, Chonnam National Univ., Information & Telecommunication Research Institute)
  • 투고 : 2012.02.21
  • 심사 : 2012.06.18
  • 발행 : 2012.07.30

초록

'제스처'는 음성을 제외한 가장 직관적인 인간의 의사표현 수단이다. 따라서 키보드나 마우스를 대체하여 제스처를 입력으로 컴퓨터를 제어할 수 있는 방법에 대한 연구가 많이 진행되고 있다. 이러한 연구에서 사용자 객체의 검출과 주요 신체부위의 추정은 매우 중요한 과정 중의 하나이다. 본 논문에서는 깊이정보가 부정확한 조건에서 사용자 객체검출과 주요 신체부위를 추정하는 방법을 제시한다. 본 논문에서는 2D 영상정보와 3D 깊이정보를 이용하여 조명 변화와 잡음에 강인하고, 3D 깊이정보를 1D 신호로 변환하여 처리함으로써 실시간에 적합하며, 이전 객체정보를 이용하여 더욱 정확하고 환경변화에 강인한 사용자 검출 방법을 제안한다. 또한 주요 신체부위 추정 방법에서 본 논문에서는 2D 외곽선 정보와 3D 깊이정보 및 추적을 혼합 사용하여 사용자 자세를 추정하는 방법을 제안한다. 실험결과 제안된 사용자 객체 검출방법은 2D정보만을 이용하는 방법에 비해 조명변화와 복잡한 환경에 강인하고, 깊이정보가 부정확한 경우에도 정확한 객체검출을 수행하였다. 또한 제안된 주요 신체부위 추정방법은 2D 외곽선 정보만 이용할 경우 겹친 부분에 대한 검출이 불가능하고, 색상 정보를 사용하는 방법은 조명이나 환경에 민감한 단점을 극복함을 확인할 수 있다.

'Gesture' is the most intuitive means of communication except the voice. Therefore, there are many researches for method that controls computer using gesture input to replace the keyboard or mouse. In these researches, the method of user detection and main body parts estimation is one of the very important process. in this paper, we propose user objects detection and main body parts estimation method on inaccurate depth information for pose estimation. we present user detection method using 2D and 3D depth information, so this method robust to changes in lighting and noise and 2D signal processing 1D signals, so mainly suitable for real-time and using the previous object information, so more accurate and robust. Also, we present main body parts estimation method using 2D contour information, 3D depth information, and tracking. The result of an experiment, proposed user detection method is more robust than only using 2D information method and exactly detect object on inaccurate depth information. Also, proposed main body parts estimation method overcome the disadvantage that can't detect main body parts in occlusion area only using 2D contour information and sensitive to changes in illumination or environment using color information.

키워드

참고문헌

  1. C. Stauffer and W.E.L. Grimson, "Adaptive background mixture models for real-time tracking" Proc. IEEE Int Conf. on Computer Vision and Pattern Recognition, pp.246-252, 1999.
  2. Haritaoglu. I, Harwood. D, Davis. L.S, "W4 : Real-time Surveillance of people and their activities", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.22, No.8, Aug.2000
  3. Haritaoglu. I, Harwood. D, Davis. L.S, "W4 : Who? When? Where? What? A Real Time System for Detecting and Tracking People", Automatic Face and Gesture Recognition,1998.IEEE, Vol, No, pp.222-227, Aug.2000
  4. Shao-Yi Chien, Yu-Wen Huang, Bing-Yu Hsieh, Shyh-Yih Ma, Liang-Gee Chen, "Fast video segmentation algorithm with shadow cancellation, global motion compensation, and adaptive threshold techniques," Multimedia, IEEE Transaction on, Vol.6, No.5, pp.732-748, Oct.2004. https://doi.org/10.1109/TMM.2004.834868
  5. Chia-Feng Juang, Chia-Ming Chang, Jiuh-Rou Rou, Lee D, "Computer vision-based human body Segmentation and posture estimation", Systems, Man and Cybernetics, Part A, Systems and Humans, IEEE Transactions on, Vol.39, No.1, pp.119-133, Jan.2009 https://doi.org/10.1109/TSMCA.2009.2008397
  6. Stella X. Yu, R. Gross and J. Shi, "Concurrent object recognition and segmentation by graph partitioning", Proc. Neural Information Processing Systems(NIPS'02), pp.1383-1390.
  7. Ismail Haritaoglu. David Harwood, Larry S. Davis, "W4S: A Real-Time System for Detecting and Tracking People in 2.5D", ECCV'98 In Computer Vision, Vol, No, pp., 1998
  8. Parvizi. E, Wu, Q.M.J, "Multiple Object Tracking Based on Adaptive Depth Segmentaion", Computer and Robot Vision 2008, Canadian Conference on, Vol, No, pp.273-277, May.2008
  9. Yinghua Shen, Chaohui Lu, Pin Xu, "Stereoscopic Video Object Segmentation Based on Disparity Map", Measuring Technology and Mechatronics Automation 2010, Vol.3, No, pp.493-495, March.2010
  10. Boykov, Y, Kolmogorov. V, "An Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimiztion in Vision", IEEE Transactions on PAMI, Vol.26, No9, pp.1124-1137, 2004 https://doi.org/10.1109/TPAMI.2004.60
  11. Fujiyoshi. H, Lipton A.J, "Real-time human motion analysis by image skeletonization" Applications of Computer Vision, 1998, WACV98, Proceedings, Fourth IEEE Workshop on, Vol., No., pp.15-21, Oct.1998
  12. Haritaoglu. I, Harwood. D, Davis L.S, "Ghost : a human body part labeling system using silhouettes", Pattern Recognition, 1998.Proceedings, Fourteenth International Conference on, Vol.1, No., pp.77-82, Aug.1998
  13. Sangho Park, J.K. Aggarwal, "Segmentation and Tracking of Interacting Human Body Parts under Occlusion and Shadowing", Proc. Motion and Video Computing(MOTION'02), pp.105-111, Dec.2002.
  14. Kikuo Jujimura, Touding Zhu, Victor Ng-Thow-Hing, "Estimating Pose from Depth Image Stream", IEEE International Conference on Humanoid Robots 2005, pp.154-160, 2005.
  15. Y. Ma, S. Worrall, A.M. Kondoz, "Automatic video object segmentation using depth information and an active contour model," Multimedia Signal Processing, 2008 IEEE 10th Workshop on, pp.910-914, Oct.2008.
  16. Comaniciu D, Meer P, "Mean shift: a robust appoach toward feature space analysis," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol, No, pp.603-619, 2002