Title :
A trial to represent dynamic concepts
Author :
Kawase, Kazushi ; Takagi, Tomohiro
Author_Institution :
Dept. of Comput. Sci., Meiji Univ., Kanagawa, Japan
Abstract :
We consider the expression and recognition of dynamic concepts by regarding the movement patterns learned in a recurrent neural net as symbols. We then develop a method to express more abstract dynamic concepts by combining them with symbols and connecting several recurrent neural networks. Application of the method to actual recognition cases, ball bouncing and dancing, demonstrated its effectiveness. These experiments show the ability of the method to deal with dynamic concepts that are difficult to describe because of vagueness.
Keywords :
image motion analysis; image recognition; image retrieval; recurrent neural nets; video signal processing; ball bouncing; dancing; dynamic concept expression; dynamic concept recognition; learned movement; recurrent neural net; symbols; vagueness; video retrieval; Computer science; Content based retrieval; Image databases; Image retrieval; Image segmentation; Indexing; Joining processes; Layout; Pattern recognition; Recurrent neural networks;
Conference_Titel :
Fuzzy Information Processing Society, 2002. Proceedings. NAFIPS. 2002 Annual Meeting of the North American
Print_ISBN :
0-7803-7461-4
DOI :
10.1109/NAFIPS.2002.1018112
