Abstract
Large networks of cameras have been increasingly employed to capture dynamic events for tasks such as surveillance and training. When using active cameras to capture events distributed throughout a large area, human control becomes impractical and unreliable. This has led to the development of automated approaches for online camera control. We introduce a new automated camera control approach that consists of a stochastic performance metric and a constrained optimization method. The metric quantifies the uncertainty in the state of multiple points on each target. It uses state-space methods with stochastic models of target dynamics and camera measurements. It can account for occlusions, accommodate requirements specific to the algorithms used to process the images, and incorporate other factors that can affect their results. The optimization explores the space of camera configurations over time under constraints associated with the cameras, the predicted target trajectories, and the image processing algorithms. The approach can be applied to conventional surveillance tasks (e.g., tracking or face recognition), as well as tasks employing more complex computer vision methods (e.g., markerless motion capture or 3D reconstruction).
- B. D. Allen. 2007. Hardware design optimization for human motion tracking systems. Ph.D. thesis, University of North Carolina at Chapel Hill. Google ScholarDigital Library
- B. D. Allen and G. Welch. 2005. A general method for comparing the expected performance of tracking and motion capture systems. In Proceedings of the 12th ACM Symposium on Virtual Reality Software and Technology. ACM, 201--210. Google ScholarDigital Library
- A. D. Bagdanov, A. Del Bimbo, and F. Pernici. 2005. Acquisition of high-resolution images through online saccade sequence planning. In Proceedings of the 3rd ACM International Workshop on Video Surveillance & Sensor Networks. 121--130. Google ScholarDigital Library
- A. Bakhtari, M. D. Naish, M. Eskandari, E. A. Croft, and B. Benhabib. 2006. Active-vision-based multisensor surveillance-an implementation. IEEE Trans. Syst. Man Cybern. Part C Appl. Rev. 36, 5, 668--680. Google ScholarDigital Library
- R. Bodor, P. Schrater, and N. Papanikolopoulos. 2005. Multi-camera positioning to optimize task observability. In Proceedings of the IEEE Conference on Advanced Video and Signal Based Surveillance. 552--557.Google Scholar
- C. Broaddus, T. Germano, N. Vandervalk, A. Divakaran, S. Wu, and H. Sawhney. 2009. Activision: Active collaborative tracking for multiple ptz cameras. In Proceedings of SPIE: Multisensor, Multisource Information Fusion: Architectures, Algorithms, and Applications. Vol. 7345.Google Scholar
- M. Casper. 2007. EventScripts Python. http://python.eventscripts.com.Google Scholar
- C. B. Chang and J. A. Tabaczyinski. 1984. Application of state estimation to target tracking. IEEE Trans. Autom. Control 29, 2, 98--109.Google ScholarCross Ref
- X. Chen. 2002. Designing multi-camera tracking systems for scalability and efficient resource allocation. Ph.D. thesis, Stanford University. Google ScholarDigital Library
- A. K. R. Chowdhury and R. Chellappa. 2004. An information theoretic criterion for evaluating the quality of 3D reconstructions from video. IEEE Trans. Image Process. 13, 7, 960--973. Google ScholarDigital Library
- R. T. Collins, A. J. Lipton, et al. 2000. A system for video surveillance and monitoring. Tech. rep. CMU-RI-TR-00-12, Robotics Institute, Carnegie Mellon University, Pittsburgh, PA.Google Scholar
- C. J. Costello, C. P. Diehl, A. Banerjee, and H. Fisher. 2004. Scheduling an active camera to observe people. In Proceedings of the ACM 2nd International Workshop on Video Surveillance & Sensor Networks. 39--45. Google ScholarDigital Library
- J. Davis. 2002. Mixed scale motion recovery. Ph.D. thesis, Stanford University. Google ScholarDigital Library
- A. J. Davison and D. W. Murray. 2002. Simultaneous localisation and map-building using active vision. IEEE Trans. Pattern Anal. Mach. Intell. Google ScholarDigital Library
- A. Del Bimbo and F. Pernici. 2005. Towards on-line saccade planning for high-resolution image sensing. Pattern Recognit. Lett. 27, 15, 1826--1834. Google ScholarDigital Library
- J. Denzler and C. Brown. 2002. An information theoretic approach to optimal sensor data selection for state estimation. IEEE Trans. Pattern Anal. Mach. Intell. 24, 2, 145--157. Google ScholarDigital Library
- J. Denzler, C. M. Brown, and H. Niemann. 2001. Optimal camera parameter selection for state estimation with applications in object recognition. In Proceedings of the 23rd DAGM-Symposium on Pattern Recognition. Lecture Notes in Computer Science, vol. 2191, 305. Google ScholarDigital Library
- J. Denzler and M. Zobel. 2001. On optimal observation models for kalman filter based tracking approaches. Tech. rep., Lehrstuhl für Mustererkennung, Universität Erlangen-Nürnberg.Google Scholar
- J. Denzler, M. Zobel, and H. Niemann. 2002. On optimal camera parameter selection in kalman filter based object tracking. In Proceedings of the 24th DAGM Symposium on Pattern Recognition. 17--25. Google ScholarDigital Library
- J. Denzler, M. Zobel, and H. Niemann. 2003. Information theoretic focal length selection for real-time active 3-d object tracking. In Proceedings of the International Conference on Computer Vision. Vol. 1, 400--407. Google ScholarDigital Library
- B. Deutsch, H. Niemann, and J. Denzler. 2005. Multi-step active object tracking with entropy based optimal actions using the sequential kalman filter. In Proceedings of the IEEE International Conference on Image Processing. Vol. 3, 105--108.Google Scholar
- B. Deutsch, S. Wenhardt, and H. Niemann. 2006. Multi-step multi-camera view planning for real-time visual object tracking. In Proceedings of the 26th DAGM-Symposium on Pattern Recognition. Lecture Notes in Computer Science, vol. 4174, 536--545. Google ScholarDigital Library
- B. Deutsch, M. Zobel, J. Denzler, and H. Niemann. 2004. Multi-step entropy based sensor control for visual object tracking. In Proceedings of the 28th DAGM-Symposium on Pattern Recognition. Lecture Notes in Computer Science, vol. 3175, 359--366.Google ScholarCross Ref
- M. S. Grewal and A. P. Andrews. 1993. Kalman Filtering Theory and Practice. Information and System Sciences Series, Prentice Hall, Upper Saddle River, NJ. Google ScholarDigital Library
- L. Guan. 2010. Multi-view dynamic scene modeling. Ph.D. thesis, University of North Carolina at Chapel Hill. Google ScholarDigital Library
- A. Ilie and G. Welch. 2011. On-line control of active camera networks for computer vision tasks. In Proceedings of the International Conference for Distributed Smart Cameras.Google Scholar
- A. Ilie, G. Welch, and M. Macenko. 2008. A stochastic quality metric for optimal control of active camera network configurations for 3D computer vision tasks. In Proceedings of the Workshop on Multi-camera and Multimodal Sensor Fusion Algorithms and Applications.Google Scholar
- D. A. Ilie. 2010. On-line control of active camera networks. Ph.D. thesis, University of North Carolina at Chapel Hill. Google ScholarDigital Library
- T. Kailath, A. H. Sayed, and B. Hassibi. 2000. Linear Estimation. Information and System Sciences Series. Prentice Hall, Upper Saddle River, NJ.Google Scholar
- N. Krahnstoever, M. Yeasin, and R. Sharma. 2001. Towards a unified framework for tracking and analysis of human motion. In Proceedings of the IEEE Workshop on Detection and Recognition of Events in Video.Google Scholar
- N. Krahnstoever, T. Yu, S.-N. Lim, K. Patwardhan, and P. Tu. 2008. Collaborative real-time control of active cameras in large scale surveillance systems. In Proceedings of the Workshop on Multi-Camera and Multi-Modal Sensor Fusion Algorithms and Applications.Google Scholar
- S.-N. Lim, A. Mittal, and L. Davis. 2005. Constructing task visibility intervals for a surveillance system. In Proceedings of the 3rd ACM International Workshop on Video Surveillance & Sensor Networks. 141--148. Google ScholarDigital Library
- S.-N. Lim, A. Mittal, and L. S. Davis. 2007. Task scheduling in large camera networks. In Proceedings of the Asian Conference on Computer Vision. Google ScholarDigital Library
- L. Marcenaro, F. Oberti, G. L. Foresti, and C. Regazzoni. 2001. Distributed architectures and logical-task decomposition in multimedia surveillance systems. Proc. IEEE 89, 10, 1419--1440.Google ScholarCross Ref
- T. Matsuyama and N. Ukita. 2002. Real-time multitarget tracking by a cooperative distributed vision system. Proc. IEEE. 90. 1137--1150.Google Scholar
- A. Mittal and L. Davis. 2004. Visibility analysis and sensor planning in dynamic environments. In Proceedings of the European Conference on Computer Vision.Google Scholar
- A. Mittal and L. S. Davis. 2008. A general method for sensor planning in multi-sensor systems: Extension to random occlusion. Int. J. Comput. Vision 76, 1, 31--52. Google ScholarDigital Library
- M. D. Naish, E. A. Croft, and B. Benhabib. 2001. Simulation-based sensing-system configuration for dynamic dispatching. In Proceedings of the IEEE International Conference on Systems, Man and Cybernetics. 2964--2969.Google Scholar
- M. D. Naish, E. A. Croft, and B. Benhabib. 2003. Coordinated dispatching of proximity sensors for the surveillance of manoeuvring targets. Rob. Comput. Integr. Manuf. 19, 3, 283--299.Google ScholarCross Ref
- P. Natarajan, T. N. Hoang, K. S. Low, and M. Kankanhalli. 2012. Decision-theoretic approach to maximizing observation of multiple targets in multi-camera surveillance. In Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems. Google ScholarDigital Library
- F. Oberti, G. Ferrari, and C. S. Regazzoni. 2001. Allocation strategies for distributed video surveillance networks. In Proceedings of the International Conference on Image Processing. Vol. 2. 415--418.Google Scholar
- G. Olague and R. Mohr. 2002. Optimal camera placement for accurate reconstruction. Pattern Recognit. 35, 4, 927--944.Google ScholarCross Ref
- F. Qureshi and D. Terzopoulos. 2007. Surveillance in virtual reality: System design and multi-camera control. In Proceedings of the Conference on Computer Vision and Pattern Recognition. 1--8.Google Scholar
- F. Qureshi and D. Terzopoulos. 2005a. Surveillance camera scheduling: a virtual vision approach. In Proceedings of the 3rd ACM International Workshop on Video Surveillance & Sensor Networks. ACM Pres, 131--140. Google ScholarDigital Library
- F. Qureshi and D. Terzopoulos. 2005b. Towards intelligent camera networks: A virtual vision approach. In Proceedings of the 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance. 177--184. Google ScholarDigital Library
- G. S. Ram, K. Ramakrishnan, P. Atrey, V. K. Singh, and M. S. Kankanhalli. 2006. A design methodology for selection and placement of sensors in multimedia surveillance systems. In Proceedings of the 4th ACM International Workshop on Video Surveillance & Sensor Networks. Google ScholarDigital Library
- P. Remagnino, A. Shihab, and G. Jones. 2003. Distributed intelligence for multi-camera visual surveillance. Pattern Recognit. 37, 4, 675--689.Google ScholarCross Ref
- E. Sommerlade and I. Reid. 2008a. Cooperative surveillance of multiple targets using mutual information. In Proceedings of the Workshop on Multi-camera and Multi-modal Sensor Fusion Algorithms and Applications.Google Scholar
- E. Sommerlade and I. Reid. 2008b. Information theoretic active scene exploration. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Google Scholar
- E. Sommerlade and I. Reid. 2008c. Information-theoretic decision making for exploration of dynamic scenes. In Proceedings of the 5th International Workshop on Attention in Cognitive Systems.Google Scholar
- E. Sommerlade and I. Reid. 2010. Probabilistic surveillance with multiple active cameras. In Proceedings of the IEEE International Conference on Robotics and Automation.Google Scholar
- P.-N. Tan, M. Steinbach, and V. Kumar. 2005. Introduction to Data Mining. Addison-Wesley. Google ScholarDigital Library
- K. A. Tarabanis, R. Y. Tsai, and P. K. Allen. 1995. A survev of sensor planning in computer vision. IEEE Trans. Rob. Autom. 11, 1, 86--104.Google ScholarCross Ref
- G. R. Taylor, A. J. Chosak, and P. C. Brewer. 2007. OVVV: Using virtual worlds to design and evaluate surveillance systems. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 1--8.Google Scholar
- G. Welch, B. D. Allen, A. Ilie, and G. Bishop. 2007. Measurement sample time optimization for human motion tracking/capture systems. In Proceedings of the Workshop on Trends and Issues in Tracking for Virtual Environments at the IEEE Virtual Reality Conference. G. Zachmann, Ed.Google Scholar
- G. Welch and G. Bishop. 2001. An introduction to the Kalman filter: SIGGRAPH 2001 course 8. In Computer Graphics SIGGRAPH 2001 Course Pack. ACM Press, Addison-Wesley Publishing Company, Los Angeles, CA.Google Scholar
- J. J. Wu, R. Sharma, and T. S. Huang. 1998. Analysis of uncertainty bounds due to quantization for three-dimensional position estimation using multiple cameras. Opt. Eng. 37, 280--292.Google ScholarCross Ref
- S. Yous, N. Ukita, and M. Kidode. 2007. An assignment scheme to control multiple pan/tilt cameras for 3d video. J. Multimedia 2, 1, 10--19.Google ScholarCross Ref
- Z. Zhang. 1999. Flexible camera calibration by viewing a plane from unknown orientations. In Proceedings of the International Conference on Computer Vision.Google ScholarCross Ref
Index Terms
- Online control of active camera networks for computer vision tasks
Recommendations
Pan-tilt-zoom camera calibration and high-resolution mosaic generation
Special issue on omnidirectional vision and camera networksIn this paper, we discuss the problem of estimating parameters of a calibration model for active pan-tilt-zoom cameras. The variation of the intrinsic parameters of each camera over its full range of zoom settings is estimated through a two step ...
A unified rectification method for single viewpoint multi-camera system
AVSS '11: Proceedings of the 2011 8th IEEE International Conference on Advanced Video and Signal Based SurveillanceStereo matching and 3D reconstruction has been studied for decades as a fundamental problem in the field of computer vision. Recent years, stereo matching and 3D reconstruction with a large field of view, especially using omnidirectional vision and ...
Kruppa equation based camera calibration from homography induced by remote plane
Camera calibration is an indispensable step in retrieving 3D metric information from 2D images. One classical self-calibration method is based on Kruppa equation derived from pairwise image correspondences. However, the calibration constraints derived ...
Comments