T. Levitt, T. Binford, G. Ettinger, and P. Gelband, "Probability-based control for computer vision," in Image Understanding Workshop, pp. 355--369, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....analysis, of course, many systems like VISIONS [3] SPAM [4] or SIGMA [5] are known which use information represented in a knowlegde base. But all these systems lack a representation of camera actions. Related work on the selection and performance of actions using Bayesian networks can be found in [6, 7]. Examples of approaches for uniform representation of actions and knowledge about the task domain are the situation calculus or decision networks, see [8, 9] for an overview or [10] for an example using Bayesian networks. We found the semantic network representation formalism particularly ....

T. Levitt, T. Binford, G. Ettinger, and P. Gelband. Probability based control for computer vision. In Proc. of DARPA Image Understanding Workshop, pages 355--369, 1989.

....an optimal camera action. In classical image analysis, of course, many knowledge based approaches like VISIONS [16] SPAM [27] or SIGMA [26] are known. But all these systems lack a representation of camera actions. Related work on the selection of actions using Bayesian networks can be found in [35, 23, 21]. In contrast, we use a semantic network for knowledge representation, because we found this kind of representation particularly suitable for the description of objects. This representation is less obvious using Bayesian networks [21] The integrated knowledge representation of camera actions and ....

T. Levitt, T. Binford, G. Ettinger, and P. Gelband. Probability based control for computer vision. In Proc. of DARPA Image Understanding Workshop, pages 355--369, 1989.

....mapping knowledge to computational constraints resides in: 1) how explicit contextual knowledge can be represented as distributed implicit parameter sets, and (2) what computational mechanisms are required for effective distribution and invocation of the parameters. Early studies by Levitt et al. [16] suggested the use of Bayesian networks for knowledge representation. Recent work by Murino et al. [18] further exploited such techniques for using knowledge in the control of camera operations. For most constrained environments, we believe that Bayesian nets and associated belief revision ....

T.S. Levitt, T.O. Binford, and et al. "Probability based Control for Computer Vision". In DARPA Image Understanding Workshop, Palo Alto, California, U.S., May 1989.

....Levitt has been involved with several inter related efforts, which together constitute the largest existing effort to build a computer vision system using Bayes nets. The main problem addressed is identifying formations of military vehicles in extremely noisy overhead radar images [Levitt, 1986; Levitt et al. 1989] It is fundamentally a grouping problem. Besides constructing a complete system, which will likely be used in field trials, the primary technical contribution is the construction of a hypothesize verify control structure with dynamic instantiation of nodes and nets. The key to the success of ....

....are discussed in Section 3.3 using influence diagrams. 3.1. 2 The Composite Net A composite net uses four kinds of knowledge structured into separate nets: The PART OF net and IS A tree [Chou and Brown, 1990] are standard, and these kinds of Bayes nets were used in the vision system described in [Levitt et al. 1989]. The expected area net and task net presented here are new, as is the composite net that results from linking the four. The composite net plays an important role in TEA 1, combining all the represented knowledge into a single structure, while representing different types of knowledge in a modular ....

[Article contains additional citation context not shown here]

T. Levitt, T. Binford, G. Ettinger, and P. Gelband, "Probabilitybased Control for Computer Vision," In Proceedings: DARPA Image Understanding Workshop, pages 355--369, 1989.

....problem and the information needed for its answer, one can simulate the contributions of an information source to this answer and thus rank the actions to be performed by the system. Examples of using Bayesian networks for an active approach to computer vision are given in (Rimey and Brown 1994; Levitt, Binford, Ettinger, and Gelband 1989). 4.2.2 A Bayesian network model for multispectral and multitemporal field classification In the framework of Bayesian classification the task of assigning a field to one of N crop classes, C 1 ; CN , given a set of measurements from n sensors, X k ; k 2 f1; ng, is represented as P (C j ....

Levitt, T., T. Binford, G. Ettinger, and P. Gelband (1989). Probability-based control for computer vision. In Image Understanding Workshop (Palo Alto, CA, May 23-26,), pp. 355--369. Morgan Kaufmann, San Mateo.

....with only a few assumptions. 2 Related Work 2. 1 Bayesian Networks in Image Understanding Bayesian networks, in combination with geometric reasoning systems, have been used in 3 D object understanding to relate model components to predicted appearances and to control the interpretation process ([16], 3] 15] 2] 19] 18] 23] The highest level of the model is a hierarchy of 3 D components, which predict the appearance of surfaces and contours. These surfaces and contours in turn predict the appearance of edges. Matching edges, finally, form the evidence nodes. Another application is ....

T.S. Levitt, T.O. Binford, G.J. Ettinger, and P. Gelband. Probability-based control for computer vision. In DARPA89, pages 355--369, 1989.

No context found.

T. Levitt, T. Binford, G. Ettinger, and P. Gelband, "Probability-based control for computer vision," in Image Understanding Workshop, pp. 355--369, 1989.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC