Abstract
A computer vision application can be defined as a sequence of image processing, feature extraction, and interpretation operations that are used to solve a specific task. In most of the computer vision systems this sequence is pre-defined and static. The work presented here shows a dynamic technique for algorithm selection based on both the value of information of each operation, and the computational complexity of the operations, which is modeled in terms of cost. This technique is used in the Ascender II system to select visual operators to perform aerial image interpretation of urban regions. The results show that the cost and information driven method presented here leads to performance gains. Moreover the hierarchical structure of the system simplifies the addition of new visual operations.
Funded by the National Council for Scientific Research-CNPq, Brazil, grant number 260185/92.2, by DARPA under contract number DACA76-97-k-0005, and by the Army Research Office, under contract numbers DAAG55-97-1-0188, and DAAG55-97-1-0026.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Fink, E.: How to solve it automatically, selection among problem-solving methods. In: Proc. of the 4th Int. Conf. on Artificial Intelligence Planning Systems, pp. 128–136. AAAI Press, Menlo Park (1998)
Lobjois, L., Lemaitre, M.: Branch and Bound algorithm selection by performance prediction. In: Proc. of the 15th National Conference on Artificial Intelligence, CA, pp. 353–358 (1998)
Brooks, R.A.: Symbolic reasoning among 3-d models and 2-d images. Artificial Intelligence 17, 285–348 (1981)
Rimey, R., Brown, C.: Task-oriented vision with multiple bayes nets. In: Blake, A., Yuille, A. (eds.) Active Vision, The MIT Press, Cambridge (1992)
Mann, W.B., Binford, T.O.: An example of 3D interpretation of images using bayesian networks. In: DARPA Image Understanding Workshop, pp. 793–801 (1992)
Kumar, V.P., Desai, U.B.: Image interpretation using bayesian networks. IEEE Transactions on Pattern Analysis and Machine Intelligence 18(1), 74–77 (1996)
Shaaban, K.M., Schalkoff, R.J.: Image processing and computer vision algorithm selection and refinement using an operator assited meta algorithm. In: SPIE Proc. Visual Information Processing IV, Orlando, FL, vol. 2488, pp. 77–87 (1995)
Potkonjak, M., Rabaey, J.: Algorithm selection: a quantitative computation intensive optimization approach. In: IEEE/ACM Int. Conf. on Computer Aided Design, CA (1994)
Lagoudakis, M.G., Littman, M.L.: Algorithm selection using reinforcement learning. In: Proc. of the 17th National Conf. on Artificial Intelligence AAAI 2000, TX, p. 108 (2000)
Marengoni, M., Jaynes, C., Hanson, A., Riseman, E.: Ascender II, a visual framework for 3D reconstruction. In: Proc. of the Int. Conf. on Vision Systems, Las Palmas, Spain (1999)
Marengoni, M., Hanson, A., Zilberstein, S., Riseman, E.: Control in a 3D reconstruction system using selective perception. In: Proceedings of the International Conference in Computer Vision, Greece (1999)
Andersen, S.K., Olesen, K.G., Jensen, F.V., Jensen, F.: HUGIN – a shell for building bayesian belief universes for expert systems. In: Proceedings of the 11th International Congress on Uncertain Artificial Intelligence, pp. 1080–1085 (1989)
Lindley, D.V.: Making Decisions, 2nd edn. John Wiley and Sons, Chichester (1985)
Marengoni, M., Hanson, A., Zilberstein, S., Riseman, E.: Decision making and Uncertainty Management in a 3D reconstruction system. IEEE Transactions on Pattern Analysis and Machine Intelligence 25(7), 852–858 (2003)
International Society for Photogrammetry and Remote Sensing (ISPRS): ftp.vislist.com/IMAGERY/PTERRAIN
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Marengoni, M., Hanson, A., Zilberstein, S., Riseman, E. (2004). Cost and Information-Driven Algorithm Selection for Vision Systems . In: Campilho, A., Kamel, M. (eds) Image Analysis and Recognition. ICIAR 2004. Lecture Notes in Computer Science, vol 3211. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30125-7_65
Download citation
DOI: https://doi.org/10.1007/978-3-540-30125-7_65
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23223-0
Online ISBN: 978-3-540-30125-7
eBook Packages: Springer Book Archive