ABSTRACT
Intensive computations required for sensing and processing perceptual information can impose significant burdens on personal computer systems. We explore several policies for selective perception in SEER, a multimodal system for recognizing office activity that relies on a layered Hidden Markov Model representation. We review our efforts to employ expected-value-of-information (EVI) computations to limit sensing and analysis in a context-sensitive manner. We discuss an implementation of a one-step myopic EVI analysis and compare the results of using the myopic EVI with a heuristic sensing policy that makes observations at different frequencies. Both policies are then compared to a random perception policy, where sensors are selected at random. Finally, we discuss the sensitivity of ideal perceptual actions to preferences encoded in utility models about information value and the cost of sensing.
- M. Ben-Bassat. Myopic policies in sequential classification. IEEE Trans. Comput., 27:170--178, 1978.Google ScholarDigital Library
- R. Bolles. Verification vision for programmable assembly. In Proc. IJCAI'77, pages 569--575, 1977.Google ScholarDigital Library
- M. Brand and V. Kettnaker. Discovery and segmentation of activities in video. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(8), 2000. Google ScholarDigital Library
- M. Brand, N. Oliver, and A. Pentland. Coupled hidden markov models for complex action recognition. In Proc. of CVPR97, pages 994--999, 1996. Google ScholarDigital Library
- H. Buxton and S. Gong. Advanced Visual Surveillance using Bayesian Networks. In International Conference on Computer Vision, pages 111--123, Cambridge, Massachusetts, June 1995.Google Scholar
- J. Feldman and R. Sproull. Decision theory and artificial intelligence ii: The hungry monkey. Cognitive Science, 1:158--192, 1977.Google ScholarCross Ref
- J. Fernyhough, A. Cohn, and D. Hogg. Building qualitative event models automatically from visual input. In ICCV'98, pages 350--355, 1998. Google ScholarDigital Library
- J. Forbes, T. Huang, K. Kanazawa, and S. Russell. The batmobile: Towards a bayesian automated taxi. In Proc. Fourteenth International Joint Conference on Artificial Intelligence, IJCAI'95, 1995. Google ScholarDigital Library
- A. Galata, N. Johnson, and D. Hogg. Learning variable length markov models of behaviour. International Journal on Computer Vision, IJCV, pages 398--413, 2001. Google ScholarDigital Library
- J. Garvey. Perceptual strategies for purposive vision. Technical Report 117, SRI International, 1976.Google Scholar
- G. Gorry and G. Barnett. Experience with a model of sequential diagnosis. Computers and Biomedical Research, 1:490--507, 1968.Google ScholarCross Ref
- D. Heckerman, E. Horvitz, and B. Middleton. A nonmyopic approximation for value of information. In Proc. Seventh Conf. on Uncertainty in Artificial Intelligence, 1991. Google ScholarDigital Library
- J. Hoey. Hierarchical unsupervised learning of facial expression categories. In Proc. ICCV Workshop on Detection and Recognition of Events in Video, Vancouver, Canada, July 2001.Google ScholarCross Ref
- E. Horvitz, J. Breese, and M. Henrion. Decision theory in expert systems and artificial intelligence. International Journal of Approximate Reasoning, Special Issue on Uncertain Reasoning, 2:247--302, 1988. Google ScholarDigital Library
- E. Horvitz, A. Jacobs, and D. Hovel. Attention-sensitive alerting. In Proc. of Conf. on Uncertainty in Artificial Intelligence, UAI'99, pages 305--313, 1999. Google ScholarDigital Library
- R. Howard. Value of information lotteries. IEEE Trans. on Systems, Science and Cybernetics, SSC-3, 1:54--60, 1967.Google ScholarCross Ref
- S. S. Intille and A. F. Bobick. A framework for recognizing multi-agent action from visual evidence. In AAAI/IAAI'99, pages 518--525, 1999. Google ScholarDigital Library
- Y. Ivanov and A. Bobick. Recognition of visual activities and interactions by stochastic parsing. IEEE Trans. on Pattern Analysis and Machine Intelligence, TPAMI, 22(8):852--872, 2000. Google ScholarDigital Library
- S. Li, X. Zou, Y. Hu, Z. Zhang, S. Yan, X. Peng, L. Huang, and H. Zhang. Real-time multi-view face detection, tracking, pose estimation, alignment, and recognition, 2001.Google Scholar
- A. Madabhushi and J. Aggarwal. A bayesian approach to human activity recognition. In In Proc. of the 2nd International Workshop on Visual Surveillance, pages 25--30, 1999. Google ScholarDigital Library
- N. Oliver, E. Horvitz, and A. Garg. Layered representations for human activity recognition. In Proc. of Int. Conf. on Multimodal Interfaces, pages 3--8, 2002. Google ScholarDigital Library
- L. Rabiner and B. Huang. Fundamentals of Speech Recognition. 1993. Google ScholarDigital Library
- L. R. Rabiner. A tutorial on hidden Markov models and selected applications in speech recognition. Proceed. of the IEEE, 77(2):257--286, Feb. 1989.Google ScholarCross Ref
- R. D. Rimey. Control of selective perception using bayes nets and decision theory. Technical Report TR468, 1993. Google ScholarDigital Library
- F. B. S. Hongeng and R. Nevatia. Representation and optimal recognition of human activities. In Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR'00, 2000.Google Scholar
- T. Starner and A. Pentland. Real-time american sign language recognition from video using hidden markov models. In Proceed. of SCV'95, pages 265--270, 1995. Google ScholarDigital Library
- L. van der Gaag and M. Wessels. Selective evidence gathering for diagnostic belief networks. AISB Quarterly, (86):23--34, 1993.Google Scholar
- A. Wilson and A. Bobick. Recognition and interpretation of parametric gesture. In Proc. of International Conference on Computer Vision, ICCV'98, pages 329--336, 1998. Google ScholarDigital Library
- H. Wu and A. Cameron. A bayesian decision theoretic approach for adaptive goal-directed sensing. ICCV, 90:563--567, 1990.Google Scholar
Index Terms
- Selective perception policies for guiding sensing and computation in multimodal systems: a comparative analysis
Recommendations
Selective perception policies for guiding sensing and computation in multimodal systems: A comparative analysis
Special issue: Attention and performance in computer visionIntensive computations required for sensing and processing perceptual information can impose significant burdens on personal computer systems. We explore the use of value of information to guide sensing and analysis in automated behavior recognition ...
PriPoCoG: Guiding Policy Authors to Define GDPR-Compliant Privacy Policies
Trust, Privacy and Security in Digital BusinessAbstractThe General Data Protection Regulation (GDPR) makes the creation of compliant privacy policies a complex process. Our goal is to support policy authors during the creation of privacy policies, by providing them feedback on the privacy policy they ...
APES: Attentively Perceiving Robot
Robot vision systems--inspired by human-like vision--are required to employ mechanisms similar to those that have proven to be crucial in human visual performance. One of these mechanisms is attentive perception. Findings from vision science research ...
Comments