Complex Visual Activity Recognition Using a Temporally Ordered Database
We propose using a temporally ordered database for complex visual activity recognition. We use a temporal precedence relation together with the assumption of fixed bounded temporal uncertainty of occurrence time of an atomic activity and comparatively large temporal extent of the complex activity. Under these conditions we identify the temporal structure of complex activities as a semiorder and design a database that has semiorder as its data model. A query algebra is then defined for this data model.
KeywordsOrder Relation Activity Recognition Query Language Atomic Activity Node Label
Unable to display preview. Download preview PDF.
- 1.A. Del Bimbo, E. Vicario, and D. ZIngoni. Symbolic description and visual querying of image sequences using spatio-temporal logic. IEEE Transactions on Knowledge and Data Engineering, 7(4), 1994.Google Scholar
- 2.S. Grumbach and T. Milo. An algebra for pomsets. In ICDT’ 95, pages 191–207. Springer-Verlag, 1995.Google Scholar
- 3.Amarnath Gupta, Shailendra Bhonsle, Simone Santini, and Ramesh Jain. An event management architecture for activity recognition in a multistream video database. In Proceedings of the 1998 Image Understanding Workshop, Monterey, CA, November 1998.Google Scholar
- 4.Ivana Mikic, Simone Santini, and Ramesh Jain. Video processing and integration from multiple cameras. In Proceedings of the 1998 Image Understanding Workshop, Monterey, CA, November 1998.Google Scholar
- 5.R. H. Mohring. Computationally tractable classes of ordered sets. In I. Rival, editor, Algorithms and Order. Kluwer Academic, 1989.Google Scholar
- 6.Wilfred Ng and Mark Levine. An extension of SQL to support ordered domains in relational databases. In Proceedings of the 1997 International Database Engineering and Applications Symposium, Montreal, Que., Canada, 25–27 Aug., pages 358–367, 1997.Google Scholar
- 7.V. R. Pratt. Modelling concurrency with partial orders. International Journal of Parallel Programming, 15(1), 1986.Google Scholar
- 8.Nuno Vasconcelos and Andrew Lippman. Towards semantically meaningful feature spaces for the characterization of video content. In Proceedings of International Conference on Image Processing, Santa Barbara, CA, USA, 26–29 Oct., pages 25–28, 1997.Google Scholar