Selective perception policies for guiding sensing and computation in multimodal systems
- 5 November 2003
- proceedings article
- Published by Association for Computing Machinery (ACM)
- Vol. 100, 36-43
- https://doi.org/10.1145/958432.958442
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.Keywords
This publication has 11 references indexed in Scilit:
- Hierarchical unsupervised learning of facial expression categoriesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Learning Variable-Length Markov Models of BehaviorComputer Vision and Image Understanding, 2001
- Recognition of visual activities and interactions by stochastic parsingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2000
- Discovery and segmentation of activities in videoPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2000
- A tutorial on hidden Markov models and selected applications in speech recognitionProceedings of the IEEE, 1989
- Decision theory in expert systems and artificial intelligenceInternational Journal of Approximate Reasoning, 1988
- Myopic Policies in Sequential ClassificationIEEE Transactions on Computers, 1978
- Decision Theory and Artificial Intelligence II: The Hungry Monkey*Cognitive Science, 1977
- Experience with a model of sequential diagnosisComputers and Biomedical Research, 1968
- Value of Information LotteriesIEEE Transactions on Systems Science and Cybernetics, 1967