Camera Motion Estimation using Geometrically Symmetric Points in Subsequent Video Frames

인접 영상 프레임에서 기하학적 대칭점을 이용한 카메라 움직임 추정

  • 전대성 (대구미래대학 멀티미디어정보과학과) ;
  • 문성헌 (대구과학대학 컴퓨터공학과) ;
  • 박준호 (경운대학교 컴퓨터공학과) ;
  • 윤영우 (영남대학교 전자정보공학부)
  • Published : 2002.03.01

Abstract

The translation and the rotation of camera occur global motion which affects all over the frame in video sequence. With the video sequences containing global motion, it is practically impossible to extract exact video objects and to calculate genuine object motions. Therefore, high compression ratio cannot be achieved due to the large motion vectors. This problem can be solved when the global motion compensated frames are used. The existing camera motion estimation methods for global motion compensation have a large amount of computations in common. In this paper, we propose a simple global motion estimation algorithm that consists of linear equations without any repetition. The algorithm uses information .of symmetric points in the frame of the video sequence. The discriminant conditions to distinguish regions belonging to distant view from foreground in the frame are presented. Only for the distant view satisfying the discriminant conditions, the linear equations for the panning, tilting, and zooming parameters are applied. From the experimental results using the MPEG test sequences, we can confirm that the proposed algorithm estimates correct global motion parameters. Moreover the real-time capability of the proposed technique can be applicable to many MPEG-4 and MPEG-7 related areas.

카메라의 이동과 회전은 영상 프레임 전체에 영향을 미치는 전역 움직임(global motion)을 유발한다. 이러한 전역 움직임을 포함하는 영상을 부호화하는 경우, 변화성분 검출(change detection) 기법을 사용하여 정확한 오브젝트를 분할하는 것은 실제적으로는 불가능하며 큰 움직임 벡터로 인해 높은 압축률을 얻기 어렵다. 이러한 문제는 전역 움직임이 보상된 영상 시퀀스를 사용함으로써 해결할 수 있다. 전역 움직임 보상을 위한 기존의 카메라 움직임 추정 방법들은 계산량이 많다는 문제점을 가지고 있다. 따라서, 본 논문에서는 간단한 선 형식으로 구성되는 전역 움직임 추정 알고리즘을 제안한다 제안 알고리즘은 영상 프레임 내의 대칭점(symmetric points)의 움직임 정보를 이용하여 패닝(panning)과 틸팅(tilting), 줌잉(zooming)에 대한 전역 움직임 파라미터를 산출한다. 전역 움직임 계산에는 카메라 회전에 대해 깊이(depth)에 독립적인 원경만이 사용되며, 영상 내에서 원경을 구분하기 위한 판별식도 논문에 제시된다. 또한, MPEG 시험 영상을 사용한 실험 결과도 나타내었다. 본 논문에서 제안한 기법의 실시간 수행 능력은 많은 영상처리 분야의 전처리 단계에서 사용될 수 있다.

Keywords

References

  1. ISO/IEC/ JTCL/SC29/WG11 N2459, 'Overview of the MPEG-4 Standard', Atlantic City, Oct. 1998
  2. ISO/IEC JTCI/SC29/WG11 2702, 'Description of automatic segmentation tecniques developed and tested for MPEG-4 versionl,' Oct. 1997
  3. R. Mech, M. Wollborn, 'A Noise Robust Method for Segmentation of Moving Objects in Video Sequences,' Proceedings of ICASSP'97, pp. 2657-2660, 1997 https://doi.org/10.1109/ICASSP.1997.595335
  4. W. Zheng, Y. Shishikui, Y. Kanatsugu, Y. Tanaka, 'A High-Precision Camera Operation Parameter Measurement System and Its Application to Image Motion Inferring,' Proc. IEEE Intl Conf. Image Processing, Vol. 3, Kobe, Japan, pp. 695-699, Oct. 1999 https://doi.org/10.1109/ICIP.1999.817205
  5. R. Wang, T. Huang, 'Fast Camera Motion Analysis in MPEG Domain', Proc. IEEE Intl Conf. Image Processing, Vol. 3, Kobe, Japan, pp. 691-694, Oct. 1999 https://doi.org/10.1109/ICIP.1999.817204
  6. 장석우, 이근수, 최형일, '어파인 모델을 이용한 카메라의 동작 추출', 정보과학회논문지, Vol. 26, No. 8(B), pp. 1000-1009, Aug. 1999
  7. G .B. Rath, A. Makur, 'Iterative Least Squares and Compression Based Estimation for a Four-Parameter Linear Global Motion Model and Global Motion Compensation,' IEEE Transactions on Circuits and Systems for Video Technology, Vol. 9, No. 7, pp. 1075-1099, Oct. 1999 https://doi.org/10.1109/76.795060
  8. A. Akutsu, Y.Tonomura, 'ACM Multimedia 94,Video Tomography: An Efficient Method for Camerawork Extraction and Motion Analysis,' ACM Multimedia 94, pp. 349-356, 1994 https://doi.org/10.1145/192593.192697
  9. F. Idris, S. Panchanathan, 'Detection of Camera Operations in Compressed Video Sequences,' Proceedings of the SPIE-Storage and Retrieval for Image and Video Database V, Vol. 3022, pp. 493-505, 1997 https://doi.org/10.1117/12.263439
  10. M. Hoetter, 'Differential estimation of the global motion parameters zoom and pan,' Signal Processing, Vol.16, pp. 249-265, 1989 https://doi.org/10.1016/0165-1684(89)90133-3
  11. Y. T. Tse, R. L. Baker, 'Global zoom/pan estimation and compensation for video compression,' Proceedings of ICASSP'91, pp. 2725-2728, 1991 https://doi.org/10.1109/ICASSP.1991.150965
  12. K. Kamikura, H. Watanabe, 'Global motion compensation in video coding,' Electronics Communications, Japan, Vol. 78, No. 4, pp. 91-102, 1995 https://doi.org/10.1002/ecja.4410780409
  13. Y.-P. Tan, S. R. Kulkami, P. J. Ramadge, 'A new method for camera motion parameter estimation,' Proc. Int. Conf. Image Processing ICIP, Vol. 1, pp. 405-408, 1995 https://doi.org/10.1109/ICIP.1995.529732
  14. K. Illgner, C. Stiller, F. Muller, 'A robust zoom and pan estimation technique,' Proc. Int. Picture Coding Symposium, PCS'93, pp. 10.5-10.6, March 1993
  15. A. Zakhor, F. Lari, 'Edge-based 3-D camera motion estimation with application to video coding,' IEEE Tr. Image Processing, Vol. 2, pp. 481-498, Oct. 1993 https://doi.org/10.1109/83.242357
  16. F. Moscheni, F. Dufaux, M. Kunt, 'A new two-stage global/local motion estimation based on a background/foreground segmen-tation,' Proc. ICASSP, pp. 2261-2264, 1995
  17. J. Heuer, A. Kaup, 'Global Motion Estimation in Image Sequences Using Robust Motion Vector Field Segmentation', Proceedings of the 7th ACM International Conference on Multimedia, pp. 261-264, 1999 https://doi.org/10.1145/319463.319620
  18. A. Akutsu, Y. Tonomura, 'Video Tomography: An efficient method for Camerawork Extraction and Motion Analysis', Proceedings of the 2nd ACM International Conference on Multimedia, pp. 349-356, 1994 https://doi.org/10.1145/192593.192697