Abstract
Calibration is fundamental, but difficult problem in computer vision. It is even difficult to calibrate an active vision system where the head geometry and/or camera intrinsic parameters change. Conventionally, calibration aims to calibrate a global model for various tasks, and this is difficult, if not impossible, for an active vision system. We propose that calibration be task-related with different models for different tasks. This simplifies the calibration process yet yields sufficient accuracy for the related task. A general model for computing the ground plane transformation for a common elevation active stereo head is derived and an approach to identifying the related parameters using real image data is given. We demonstrate that after compensating for manufacturing errors, this general ground plane transformation model can be used in a real system to compute the ground plane transformation in real-time from according to the head feedback state.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
R. Deriche, Z. Zhang, Q. T. Luoug, and O. Faugeras. Robust recovery of the epipolar geometry for an uncalibrated stereo rig. In Proc. 3rd European Conference on Computer Vision, pages 567–576. Springer-Verlag, 1994.
F. Du and J. M. Brady. A four degree-of-freedom robot head for active vision. International Journal of Pattern Recognition and Artificial Intelligence, 8(6):1439–1469, 1994.
F. Du and M. Brady. Self-calibration of the intrinsic parameters of cameras for active vision systems. IEEE Conf. on Computer Vision and Pattern Recognition, pages 477–482, 1993.
F. Ferrari, E. Grosso, G. Sandini, and M. Magrassi. A stereo vision system for real time obstacle avoidance in unknown environment. IEEE International Workshop on Intelligent Robots and Systems, 2:703–8, 1990.
E. Krotkov. Active computer vision by cooperative focus and stereo. Springer Series in Perception Engineering, 1989.
F. Li. Active Stereo for AGV Navigation. PhD thesis, University of Oxford, 1996.
F. Li, J. M. Brady, and C. Wiles. Fast computation of the fundamental matrix for an active stereo vision system. In Proc. of ECCV'96, pages (I) 157–166, Cambridge, England, 1996. Springer-Verlag.
M. Li. Camera calibration of the KTH head/eye system. Technical report, CVAP147, Computational Vision and Active Perception Laboratory, Royal Institute of Technology, S-100 44,Stockholm, Sweden, February, 1994.
S. D. Ma. A self-calibration of technique for active vision systems. IEEE Transactions on Robotics and Automation, 12(1):114–120, 1996.
P. F. McLauchlan and D. W. Murray. Active camera calibration for a head-eye platform using the variable state-dimension filter. IEEE Transactions on Pattern Analysis and Machine Intelligence, 18(1):15–22, 1996.
I.D. Reid and P. A. Beardsley. Self-alignment of a binocular robot. In Proc. 6th British Machine Vision Conference, Birmingham, 1995.
R. Y. Tsai. An efficient and accurate camera calibration technique for 3d machine vision. In Proc. of IEEE International Conference on Computer Vision and Pattern Recognition, pages 364–374, Miami, FL, June 22–26, 1986.
G-Q. Wei and S-D. Ma. Implicit and explicit camera calibration: Theory and experiments. IEEE Transactions on Pattern Analysis and Machine Intelligence, 16(5):469–480, 1994.
M. Xie. Ground plane obstacle detection from stereo pairs of images without matching. In Proc. ACCV'95, pages II-280, Singapore, 1995.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Li, F., Brady, M. (1997). Dynamic calibration of an active vision system to compute the ground plane transformation. In: Chin, R., Pong, TC. (eds) Computer Vision — ACCV'98. ACCV 1998. Lecture Notes in Computer Science, vol 1351. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63930-6_112
Download citation
DOI: https://doi.org/10.1007/3-540-63930-6_112
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63930-5
Online ISBN: 978-3-540-69669-8
eBook Packages: Springer Book Archive